A Theory of Governmental Damages Liability - Torts Constitutional

A THEORY OF GOVERNMENTAL DAMAGES LIABILITY:
TORTS, CONSTITUTIONAL TORTS, AND TAKINGS
Lawrence Rosenthat
TABLE OF CONTENTS
INTRO DU CTIO N .................................................................................... 798

I. THE CONTOURS OF GOVERNMENTAL TORT LIABILITY .................... 800
A. Com m on-Law Torts .......................................................... 801
1. Federal Liability .................................................... 801
2. State and Local Liability ....................................... 804
3. Asymmetry with Private Tort Liability .................. 813
B. Constitutional Torts .......................................................... 815
1. Federal Liability .................................................... 815
2. State and Local Liability ....................................... 816
3. Asymmetry with Private Tort Liability .................. 818
C . T akin gs .............................................................................. 821
II. THE POLITICS OF GOVERNMENTAL TORT LIABILITY ...................... 822
A. The Case Against Governmental Tort Liability ............... 822
1. Theories of Tort Liability ..................................... 822
2. The Inapplicability of the Conventional Theory to
G overnm ent ........................................................ 824
B. The Politics of Governmental Damages Liability ............ 831
1. A Theory of Political Behavior ............................. 832
2. Assessing the Theory ............................................. 833
3. Immunity Legislation as Empirical Evidence in
Support of the Theory ....................................... 838
4. Corroborative Evidence for the Theory ............... 840
C. The Impact of Governmental Tort Liability .................... 842
III. ASSESSING GOVERNMENTAL LIABILITY .......................................... 844
A. Com mon-Law Torts .......................................................... 847
1. The Marginal Utility of Governmental Liability.. 847
2. The Difficulty of Adjudicating Governmental
L iability ............................................................... 850
3. The Weak Case for Governmental Liability ........ 853
B. Constitutional Torts .......................................................... 856
1. The Irrelevance of Political Accountability ......... 856
2. Qualified Immunity and Vicarious Employer
JOURNAL OF CONSTITUTIONAL LA W [Vol. 9:3
L iab ility ............................................................... 856

3. M easure of Dam ages ............................................. 860
4. Statutory Limitations on Damages ....................... 861
C . Takin gs .............................................................................. 863
1. The Role of Compensation in Limiting Overuse of
Em inent Dom ain ................................................ 863
2. The Public-Use Inquiry ......................................... 864
3. The Standard forJust Compensation .................. 868
C O N CLU SIO N ........................................................................................ 870
INTRODUCTION
Governmental damages liability in tort represents a special case, to

say the least. The rules governing governmental liability are riddled
with immunities unknown in the private sector-a confusing patch-
work seemingly without explanation.1 Governmental tort liability is
also without ajustificatory theory. Theories of tort liability generally
fall within two broad camps: the instrumentalists claim that tort li-
ability promotes efficient investments in safety by visiting financial
consequences on those who underinvest in safety, and the advocates
of corrective justice claim that tort liability embodies a moral obliga-
tion of culpable parties to provide compensation for losses for which
they are fairly considered responsible.2 Neither theory, however, of-
fers much support for governmental tort liability. Unlike private tort-
feasors, the government's objective is not profit maximization; it re-
sponds to political and not market discipline. Thus, the instrumental
justification for tort liability is wanting in the public sector, at least in
the economic terms used by the instrumentalists. 3 As for corrective
justice, the government passes its legal costs along to the taxpayers,
who bear little meaningful culpability for the underlying tortious
conduct, but who can be taxed to fund essentially unlimited liability
far in excess of the exposure to liability faced, for example, by a
shareholder in a private corporation. Thus, corrective justice also
Associate Professor of Law, Chapman University School of Law. I received assistance

from
many quarters as I worked on this Article, but I owe particular thanks for their incisive advice to
Tom Bell, Denis Binder, Henry Butler, Don Gifford, Donald Kochan, Celestine McConville,
Timothy Lytton, Patricia Salkin, and Scott Spitzer. I am also indebted to Kerry Franich, Re-
becca Meyer, and the Chapman Law Library staff for their highly capable research assistance.
See infra Part 1.
For explication of this bifurcated characterization of tort theory, see, for example, Chris-
topherJ. Robinette, Can There Be a Unifed Theory of Torts? A PluralistSuggestion from History and
Doctrine,43 BRANDEIS L.J. 369, 370 (2005), and Gary T. Schwartz, Mixed Theories of Tort Law: Af-
firmingBoth Deterrenceand CorrectiveJustice, 75 TEX. L. REv. 1801, 1801-11 (1997).
See infratext accompanying notes 112-21.
Feb. 2007] A THEORY OF GO VERNMENTAL DAMA GES LIABILITY 799
supplies little support for public-sector tort liability. 4 Indeed, there is

an emerging consensus among legal scholars that governmental tort
liability lacks a coherent justification. Moreover, this unease with
governmental liability is not confined to the academy; the United
States Supreme Court has itself expressed doubts about the utility of
damages awards against the government, leaving unclear what, if any,
justification might support even limited forms of governmental liabil-
ity for damages.
In the discussion that follows, I mean to show that the emerging
consensus is wrong. To do so, I anchor the justification for govern-
mental tort liability in a theory of political behavior. I look to politics
because the government responds primarily to political costs and
benefits, whereas private tortfeasors respond primarily to economic
rewards or punishment. I argue that governmental tort liability ex-
acts a political price by diverting the funds used to pay judgments and
other litigation costs from what elected officials regard as their politi-
cally optimal use. Governmental liability therefore creates a political
incentive to invest in loss prevention in order to maximize political
control over tax and spending policy. This theory, however, does not
argue for unlimited governmental liability; to the contrary, it provides
a justification for most of the immunities from governmental liability
that have been recognized.
The discussion below proceeds in three parts. Part I describes the
scope of governmental damages liability. Part II then demonstrates
the inability of conventional theories of tort law to support govern-
mental liability. It goes on to offer a justificatory theory for govern-
mental tort liability: the theory of political behavior advanced in Part
II suggests that governmental damages liability can be expected to
create a political incentive for the government to make cost-justified
investments in safety not present in a regime of nonliability. Part II
submits that there is considerable empirical evidence to support the
theory of political behavior it advances-the consistent legislative
practice of enacting governmental immunity statutes.
While Parts I and II are largely descriptive, Part III is evaluative. It
argues that the case for common-law tort liability against government
is weak because its marginal utility is unclear-there are important
political incentives to invest in loss prevention apart from the threat
of tort liability. Moreover, tort actions against the government re-
quire juries to assess the manner in which scarce public resources are
allocated among competing priorities, something jurors have little
ability to do. And forcing the government to divert scarce resources
4 See infra text accompanying notes 122-24.

to the defense of litigation and the payment of judgments has impor-

tant adverse impacts on essentially innocent third parties-the tax-
payers and those dependent upon the ability of government to ade-
quately fund public services. Part III therefore argues that
governmental tort liability is unwarranted when it is reasonable to be-
lieve that ordinary political accountability will adequately encourage
governmental investments in safety. Statutory immunity for discre-
tionary decision-making is a good example of the point-it confers
immunity in cases when political accountability will likely provide suf-
ficient protection for the public without need of a damages remedy.
Limitations on recoverable damages-such as ceilings on damages
awards and a prohibition on punitive damages-are warranted as a
means of mitigating the adverse effects of governmental liability.
When it comes to constitutional torts, however, Part III takes a dif-
ferent view. The Constitution does not leave its enforcement to the
political process; accordingly, political accountability is never an ade-
quate remedy for a constitutional violation. Discretionary and other
categorical immunities are therefore inappropriate for constitutional
violations; a law of constitutional torts should place pressure on the
government to conform all of its conduct to the Constitution. That
does not mean, however, that every constitutional violation must re-
sult in a damages award. The doctrine of qualified immunity prop-
erly limits liability when the government has committed adequate re-
sources to avoiding constitutional violations. Statutory limitations on
recoverable damages and the requirement that only traditional tort
damages be awarded are also justifiable as a means of mitigating the
risk that large damages awards will compromise the government's
ability to provide public services. But for one type of constitutional
liability rule-the obligation to pay just compensation when govern-
ment takes private property for a public purpose-immunity is never
appropriate. The compensation requirement creates a political
check against unwarranted takings; officials incur a political cost
when they must allocate public resources to the payment of compen-
sation. Still, the political costs imposed by the just-compensation re-
quirement are particularly high, and for that reason the Supreme
Court has been correct to defer to the political process on the ques-
tion of what kinds of takings are for a public use, as in the Court's re-
cent decision in Kelo v. City of New London.7
I. THE CONTOURS OF GOVERNMENTAL TORT LIABILITY
The rules for governmental tort liability are complex. One set of
rules controls suits against a unit of government, and another con-
7 545 U.S. 469 (2005).

Feb. 2007] A THEORY OF GOVERNMENTAL DAMAGES LIABILITY 801
trols suits against a public employee, even when sued for acts under-
taken within the scope of employment and at the direction of a gov-
ernmental employer. Liability rules differ as well for federal, state,
and local governments, and for constitutional and nonconstitutional
torts. One type of constitutional injury-a "taking" of private prop-
erty for public use-receives different treatment altogether."
A. Common-Law Torts
The history of governmental liability in tort reflects an evolution
from a common-law doctrine of sovereign immunity to a seemingly
unprincipled patchwork of statutory immunities.
1. FederalLiability
Under the doctrine of sovereign immunity, the federal govern-
ment is immune from liability for damages without its consent. 9 Fed-
eral employees, however, are personally liable for their own wrongful
conduct even when acting within the scope of employment under this
common-law doctrine. 0
It is sometimes said that the sovereign immunity of the United
States is inconsistent with the Constitution, which treats suits against
the United States as within the scope of the judicial power." To be
sure, Article III provides that "[t] he judicial Power shall extend... to
Controversies to which the United States shall be a Party,'02 and the
Constitution breathes not a word about a federal immunity from li-
' Governmental liability for breach of contract involves considerations beyond the scope of
this Article. For a helpful discussion of governmental contractual liability, see Daniel R. Fischel
& Alan 0. Sykes, Government Liabilityfor Breach of Contract, 1 AM. L. & ECON. REV. 313 (1999).
9 See, e.g.,United States v. Testan, 424 U.S. 392, 399 (1976); United States v. Sherwood, 312
U.S. 584, 587 (1941). For early statements of the doctrine, albeit containing little in the way of
a supporting rationale, see United States v. Clarke, 33 U.S. (8 Pet.) 436, 444 (1834), and
Cohens v. Virginia, 19 U.S. (6 Wheat.) 264, 411-12 (1821).
'0See, e.g., Brady v. Roosevelt S.S. Co., 317 U.S. 575, 580 (1943); Sloan Shipyards Corp. v.
U.S. Shipping Bd. Emergency Fleet Corp., 258 U.S. 549, 567 (1922). At common law, this rule
was not applicable, however, when the relief sought went beyond the payment of damages by
the individual defendant and involved remedial measures thought to implicate the interests of
the sovereign. See, e.g., Larson v. Domestic & Foreign Commerce Corp., 337 U.S. 682, 687-88
(1949). For general discussions of the scope of federal sovereign immunity, see, for example,
David P. Currie, Sovereign Immunity and Suits Against Government Officers, 1984 SUP. CT. REV. 149;
David E. Engdahl, Immunity and Accountabilityfor Positive Government Wrongs, 44 U. COLO. L. REV.
1 (1972); Vicki C. Jackson, Suing the Federal Government: Sovereignty, Immunity, andJudicialInde-
pendence, 35 CEO. WASH. INT'L L. REV. 521 (2003); and Ann Woolhandler, Patternsof Official Im-
munity and Accountability, 37 CASE W. RES. L. REV.396 (1987).
" See, e.g.,Jackson, supra notel0, at 523; Susan Randall, Sovereign Immunity and the Uses of His-
tory, 81 NEB. L. REV. 1, 38-46, 65-66 (2002).
12 U.S. CONST. art. III, §
2.
JOURNAL OF CONSTITUTIONAL LAW [Vol. 9:3
ability.1 3 The view that sovereign immunity has no constitutional

grounding, however, overlooks the Appropriations Clause: "[n]o
money shall be drawn from the Treasury, but in Consequence of Ap-
propriations made by Law .... Most state constitutions contain
similar restrictions. 5 Under such a constitutional limitation on the
use of public funds, a court cannot hear a case asking it to compel
" The only scholarship of which I am aware to suggest some basis for federal sovereign im-
munity in the Constitution's text is Professor Nelson's claim, albeit in the context of his discus-
sion of the immunity of nonconsenting states from suit in federal court, that the term "contro-
versy" was understood at the time of Article III's adoption to exclude a suit against a sovereign
without its consent. Professor Nelson assembles considerable evidence that in the eighteenth
century a court could not exercise jurisdiction over a sovereign without its consent but much
less evidence to support his claim that such a suit was not considered a "controversy." Caleb
Nelson, Sovereign Immunity as a Doctrine of PersonalJurisdiction, 115 HARV. L. REV. 1559, 1580-92
(2002). There is, moreover, ample historical evidence to the contrary. See, e.g., John J. Gib-
bons, The Eleventh Amendment and State Sovereign Immunity: A Reinterpretation,83 COLUM. L. REV.
1889, 1908-12 (1983). Indeed, if Professor Nelson is right about the original understanding of
Article 111, then surely it is remarkable that in Chisholm v. Georgia, 2 U.S. (2 DalI.) 419 (1793), all
of the Justices somehow failed to remember what the term "controversy" meant when they held
that Article III permitted suit against a nonconsenting state-including Justice Wilson, a mem-
ber of the committee that drafted Article III at the constitutional convention. See Nelson, supra,
at 1562. Even the dissenting opinion in Chisholm failed to rely on the meaning of the term
.controversy," Chisholm, 2 U.S. (2 DalI.) at 469-79 (Jay, C.J., dissenting), as did the contempora-
neous critics of that decision, on Professor Nelson's own account of that criticism, Nelson, su-
pra, at 1564-65. Perhaps some skepticism is in order about a view that depends on Professor
Nelson's greater familiarity with eighteenth-century legal concepts than that evinced by eight-
eenth-century lawyers and judges.
" U.S. CONST. art. I, § 9, cl. 7. For the leading account of the Appropriations Clause, see
Kate Stith, Congress'Powerof the Purse,97 YALE L.J. 1343 (1988).
" Forty states have parallel constitutional provisions. See ALA. CONST. art. IV, § 72; ALASKA
CONST. art. IX, § 13; ARK. CONST. art. 5, § 29; CAL. CONST. art. XVI, § 7; DEL. CONST. art. VIII,
§ 6(a); FLA. CONST. art. VII, § 1(c); GA. CONST. art. III, § IX, para. I; HAW. CONST. art. VII, § 5,
cl. 2; IDAHO CONST. art. VII, § 13; ILL. CONST. art. VIII, § 2(b); IND. CONST. art. 10, § 3; IOWA
CONST. art. III, § 24; KAN. CONST. art. 2, § 24; KY. CONST. § 230; LA. CONST. art. IIl, § 16; ME.
CONST. art. V, pt. 3d, § 4; MD. CONST. art. III, § 32; MINN. CONST. art. XI, § 1; MO. CONST. art.
III, § 36; MONT. CONST. art. VIII, § 14; NEB. CONST. art. 111-25; NEV. CONST. art. 4, § 19; N.J.
CONST. art. VIII, § II, para. 2; N.M. CONST. art. IV, § 30; N.Y. CONST. art. VII, § 7; N.C. CONST.
art. V, § 7; N.D. CONST. art. X, § 12, para. 1; OHIO CONST. art. II, § 22; OKLA. CONST. art. V-55;
OR. CONST. art. IX, § 4; PA. CONST. art. III, § 24; S.C. CONST. art. X, § 8; S.D. CONST. art. XI, § 9;
TENN. CONST. art. II, § 24; TEx. CONST. art. 8, § 6; VA. CONST. art. X, § 7; WASH. CONST. art.
VIII, § 4; W. VA. CONST. art. X, § 10-3; WIS. CONST. art. VIII, § 2; WO. CONST. art. 3, § 35. In
addition, three state constitutions expressly provide for sovereign immunity. See ALA. CONST.
art. I, § 14 ("That the State of Alabama shall never be made a defendant in any court of law or
equity."); ARK. CONST. art. 5, § 20 ("The State of Arkansas shall never be made defendant in any
of her courts."); W. VA. CONST. art. VI, § 6-35 ("The state of West Virginia shall never be made
defendant in any court of law or equity .... ."). Twenty-two grant the state legislature authority
to determine governmental liability. See ALASKA CONST. art. II, § 21; CONN. CONST. art. XI, § 4;
DEL. CONST. art. I, § 9; FLA. CONST. art. X, § 13; GA. CONST. art. I, § II, para. IX(a); ILL. CONST.
art. XIII, § 2(a); IND. CONST. art. 4, § 24; KY. CONST. § 231; LA. CONST. art. XII, § 10(c); MONT.
CONST. art. 2, § 18; NEB. CONST. art. V-22; NEV. CONST. art. 4, § 22; N.Y. CONST. art. III, § 19;
OHIO CONST. art. I, § 16, cl. 2; OKLA. CONST. art. XXIII-7; OR. CONST. art. IV, § 24; PA. CONST.
art. I, § 11; S.D. CONST. art. III, § 27; TENN. CONST. art. I, § 17; WASH. CONST. art. II, § 26; Wis.
CONST. art. IV, § 27; WO. CONST. art. 1, § 8.
Feb.2007] A THEORY OF GOVERNMENTAL DAMAGES LIABILITY 803
the government to pay a judgment absent legislative authorization for

payment; in such a case, the remedy sought is itself unconstitutional.
For example, in District of Columbia v. Eslin,16 while the cases at is-
sue were on appeal, Congress forbade payment of the judgments that
the plaintiffs had obtained, and the Supreme Court therefore held
that no court could enforce the judgments, rendering the cases non-
justiciable. 7 More recently, in Office of PersonnelManagement v. Rich-
mond,"' the Court held that a claimant could not use common-law es-
toppel principles to obtain federal disability benefits not authorized
by statute because a court-ordered payment of benefits without con-
gressional authorization would violate the Appropriations Clause.' 9
Thus, the Appropriations Clause and its state counterparts effectively
insulate the 0government from suit for damages absent legislative au-
thorization. An appropriations-based understanding of sovereign
immunity, moreover, explains why sovereign immunity poses no ob-
stacle to a suit against a public employee even when he acts at the di-
rection of the sovereign. Under the Appropriations Clause, the
United States cannot satisfy a damages judgment absent legislative
consent, but there is no similar restriction on enforcing a damages
award against a federal official's assets.
Despite the availability of sovereign immunity, Congress has been
unwilling to shield the federal treasury from all tort liability. The
Federal Tort Claims Act (FTCA) provides, "It]he United States shall
be liable.., in the same manner and to the same extent as a private
16 183 U.S. 62 (1901).

Id. at 65-66.
" 496 U.S. 414 (1990).
'9 Id. at 424-26.
To be sure, appropriations legislation may itself be challenged under a substantive consti-
tutional provision. See, e.g., Legal Servs. Corp. v. Velazquez, 531 U.S. 533, 540-49 (2001);
United States v. Lovett, 328 U.S. 303, 315 (1946). Moreover, under United States v. Klein, 80 U.S.
(13 Wall.) 128 (1871), appropriations legislation is thought to impermissibly infringe upon the
judicial power under Article III when it directs the judiciary to apply a rule of decision without
altering substantive law. See Miller v. French, 530 U.S. 327, 347-49 (2000); Robertson v. Seattle
Audubon Soc., 503 U.S. 429, 438-41 (1992). But, as the holding in Eslin demonstrates, Klein
does not invalidate legislation that merely declines to fund the judgment sought by the plaintiff.
Eslin, in turn, is consistent with contemporary Article III jurisprudence. Article III is under-
stood to prevent a court from heaing a case when the plaintiff's injury cannot be redressed by a
favorable judicial decision. See, e.g., Steel Co. v. Citizens for a Better Env't, 523 U.S. 83, 103-04
(1998); Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61 (1992); Allen v. Wright, 468 U.S.
737, 750-52 (1984); Simon v. E. Ky. Welfare Rights Org., 426 U.S. 26, 38 (1976). Thus, when
there is no appropriation available to pay the damages award sought by the plaintiff, a court
would be unable to hear the case under Article III because it could not issue an enforceable
damages award in light of the Appropriations Clause. Indeed, the Supreme Court held that the
Court of Claims could issue judgments against the United States without running afoul of Arti-
cle III's prohibition on advisory opinions only by relying on Congress's consistent practice of
appropriating funds to pay judgments of that court. See Glidden Co. v. Zdanok, 370 U.S. 530,
570 (1962) (Harlan,J., announcingjudgment).
JOURNAL OF CONSTITIIONAL LAW (Vol. 9:3
individual under like circumstances, but shall not be liable for inter-
est prior to judgment or for punitive damages."'" The FICA adds
that it supplies the exclusive remedy for torts committed by a federal
employee acting within the scope of employment except for an action
brought under the United States Constitution or a federal statute.22
The liability created by the FITCA is significantly circumscribed.
The FTCA grants immunity from liability on
[a] ny claim based upon an act or omission of an employee of the Gov-
ernment, exercising due care, in the execution of a statute or regulation,
whether or not such statute or regulation be valid, or based upon the ex-
ercise or performance or the failure to exercise or perform a discretion-
ary function or duty on the part of a federal agency or an employee of
abused. 3
the Government, whether or not the discretion involved be
The FTCA also confers immunity from liability for "[a] ny claim aris-
ing out of the loss, miscarriage, or negligent transmission of letters or
postal matter,, 24 a "claim arising in respect of the assessment or col-
lection of any tax or customs duty, or the detention of any goods,
merchandise, or other property by any officer of customs or excise or
any other law enforcement officer, ' 5'and "[a]ny claim arising out of
assault, battery, false imprisonment, false arrest, malicious prosecu-
tion, abuse of process, libel, slander, misrepresentation, deceit, or in-
.... ,26
terference with contract rights
2. State andLocal Liability

Only one state has enacted legislation providing that governmen-
tal defendants
27 are liable in tort on the same terms as private tortfea-
sors. All other states limit governmental tort liability by statute.
2' 28 U.S.C. § 2674 (2000).

" Id. § 2679(b).
2 Id. § 2680(a).
24 Id. § 2680(b).
25 Id. § 2680(c).
2'6Id. § 2680(h). The Act additionally confers immunity from liability for the fiscal opera-
tions of the Treasury or the regulation of the monetary system, id. § 2680(i), claims arising out
of the combatant activities of the military, id. § 2680(j), claims arising in a foreign country, id.
§ 2680(k), and claims arising from the activities of the Tennessee Valley Authority, the Panama
Canal Company, or from the activities of federal banks, id. § 2680(/)-(n).
" See WASH. REv. CODE ANN. §§ 4.92.090, 4.96.010 (West 2006). Even so, the Washington
Supreme Court has construed the statute to preserve common-law governmental immunity for
discretionary functions. See McCluskey v. Handorff-Sherman, 882 P.2d 157, 161-63 (Wash.
1994) (en banc). There is, however, a Washington statute granting public officials immunity
for discretionary acts or omissions, but it provides that their employer remains liable. WASH.
REV. CODEANN. § 4.24.470 (West 2005). For contrasting discussions of the experience of Wash-
ington under this regime, compare Michael Tardif & Rob McKenna, Washington State's 45-Year
Experiment in Government Liability, 29 SEATTLE U. L. REV. 1 (2005), with Debra L. Stephens &
Bryan P. Harnetiaux, The Value of Government Tort Liability: Washington State'sJourneyfrom Immu-
nity to Accountability, 30 SEATrLE U. L. REv. 35 (2006).
Many state immunity statutes confer immunities on governmental de-

fendants and public employees similar to those in the F'CA. For ex-
immunity,2 s
ample, thirty-three states recognize discretionary-function
twenty-three recognize immunity for injuries caused by reliance on
statutes or other enactments, '" twenty-three immunize the collection
of a tax,' seventeen immunize specified intentional torts of public
See ALASKA STAT. § 09.50.250(1) (2004 & Supp. 2005); id. § 09.65.070(d) (2); ARIZ. REV.
STAT. ANN. § 12-820.01 (2003); CAL. GOV'T CODE § 820.2 (West 1995); CONN. GEN. STAT. ANN.
§ 52-557n(a)(2)(B) (West 2005); DEL. CODE ANN. tit. 10, §§ 4001(1), 4011(b)(3) (1999 repl.);
GA. CODE ANN. § 50-21-24(2) (2006); HAW. REV. STAT. § 662-15(1) (1993 repl. & Supp. 2004);
IDAHO CODE ANN. § 6-904(1) (2004); 745 ILL. COMP. STAT. ANN. § 10/2-201 (West 2002); IND.
CODE ANN. § 34-13-3-3(7) (West 1999 & Supp. 2006); IOWA CODE ANN. §§ 669.14(1), 670.4(3)
(West 1998 & Supp. 2006); KAN. STAT. ANN. § 75-6104(e) (1997 & Supp. 2005); KY. REV. STAT.
ANN. § 44.073(13)(a) (LexisNexis 1997 repl. & Supp. 2005); id. § 65.2003(3)(d) (LexisNexis
2004 repl.); LA. REV. STAT. ANN. § 9:2798.1 (B) (1997); ME. REV. STAT. ANN. tit. 14, §§ 8104-B(3),
8111(1)(C) (2003); MD. CODE ANN., CTS. &JUD. PROC. § 5-507(b)(1) (LexisNexis 2002 repl.);
MASS. ANN. LAWS ch. 258, § 10(b) (LexisNexis 2004); MINN. STAT. ANN. § 3.736, subdiv. 3(b)
(West 2005); id. § 466.03, subdiv. 6 (West 2001 & Supp. 2006); MISS. CODE ANN. § 11-46-9(1) (d)
(West 1999); NEB. REV. STAT. § 13-910(2) (1997 & Supp. 2005); id. § 81-8,219(1) (2003 & Supp.
2005); NEV. REV. STAT. ANN. § 41.032(2) (LexisNexis 2002 repl.); N.H. REV. STAT. ANN. § 541-
B:19 (LexisNexis 2006 repl.); N.J. STAT. ANN. § 59:2-3(a), (c)-(d) (West 1992); N.D. CENT.
CODE §§ 32-12.1-03(3)(d), -12.2-02(3)(b) (1996 repl. & Supp. 2005); OHIO REV. CODE ANN.
§ 2744.03(A) (2), (5) (LexisNexis 2000 repl. & Supp. 2002); OKLA. STAT. ANN. tit. 51, § 155
(West 2000); OR. REV. STAT. ANN. § 30.265(3) (c) (West 1983 & Supp. 2005); 42 PA. CONS. STAT.
§§ 8524(3), 8546(3) (2002); S.C. CODE ANN. § 15-78-60(5) (2005 & Supp. 2005); TENN. CODE
ANN. § 29-20-205(1) (2000 repl.); TEX. CIV. PRAC. & REM. CODE ANN. § 101.056 (Vernon 2005);
UTAH CODE ANN. § 63-30d-301(5)(a) (2004 repl. & Supp. 2005); VT. STAT. ANN. tit. 12,
§ 5601 (e) (1) (2002); Wis. STAT. ANN. § 893.80(4) (West 2006). Other jurisdictions recognize
discretionary immunity as a common-law doctrine. See Powell v. District of Columbia, 602 A.2d
1123, 1126 (D.C. 1992); Weiss v. Fote, 167 N.E.2d 63, 65-68 (N.Y. 1960); City of Chesapeake v.
Cunningham, 604 S.E.2d 420, 426 (Va. 2004); Libercent v. Aldrich, 539 A.2d 981, 984 (Vt.
1987); McCluskey, 882 P.2d at 161-63; Parkulo v. W. Va. Bd. of Prob. & Parole, 483 S.E.2d 507,
522 (W. Va. 1996).
' See CAL. GOV'T CODE § 820.6 (West 1995); GA. CODE ANN. § 50-21-24(1) (2006); HAw.
REV. STAT. § 662-15(1) (1993 repl. & Supp. 2004); IDAHO CODE ANN. § 6-904(1) (2004); 745
ILL. COMP. STAT. ANN. § 10/2-203 (West 2002); IND. CODE ANN. § 34-13-3-3(9) (West 1999 &
Supp. 2006); IOWA CODE ANN. §§ 669.14(1), 670.4(3) (West 1998 & Supp. 2006); KAN. STAT.
ANN. § 75-6104(c) (1997 & Supp. 2005); MASS. ANN. LAWS ch. 258, § 10(a) (LexisNexis 2004);
MINN STAT. ANN. § 3.736, subdiv. 3(a) (West 2005); id. § 466.03, subdiv. 5 (West 2001 & Supp.
2006); MISS. CODE ANN. § 11-46-9(1)(b) (West 1999); MONT. CODE ANN. § 2-9-103 (2005); NEB.
REV. STAT. § 13-910(1) (1997 & Supp. 2005); id. § 81-8,219(1) (2003 & Supp. 2005); NEV. REV.
STAT. ANN. § 41.032(1) (LexisNexis 2002 repl.); N.H. REV. STAT. ANN. § 541-B:19 (LexisNexis
2006 repl.) ; N.J. STAT. ANN. § 59:3-4 (West 1992); N.D. CENT. CODE §§ 32-12.1-03(3) (a), -12.2-
02(3)(a) (1996 repl. & Supp. 2005); OKLA. STAT. ANN. tit. 51, § 155(4) (West 2000); OR. REV.
STAT. § 30.265(3)(f) (1983 & Supp. 2005); 42 PA. CONS. STAT. ANN. § 8524(2) (2002); S.C.
CODE ANN. § 15-78-60(4) (2005 & Supp. 2005); VT. STAT. ANN. tit. 12, § 5601(e)(1) (2002); W.
VA. CODE ANN. § 29-12A-5 (a) (4) (LexisNexis 2004 repl.).
'o See ARK. CODE ANN. § 19-10-204(b) (2) (B) (1998 repl. & Supp. 2003); CAL. GOVT CODE
§ 860.2 (West 1995); GA. CODE ANN. § 50-21-24(3) (2006); HAW. REV. STAT. § 662-15(2) (1993
repl. & Supp. 2004); IDAHO CODE ANN. § 6-904(A)(1) (2004); IOWA CODE ANN. §§ 669.14(2),
670.4(2) (West 1998 & Supp. 2006); KAN. STAT. ANN. § 75-6104(f) (1997 & Supp. 2005); KY.
REV. STAT. ANN. § 65.2003(2) (LexisNexis 2004 repl.); MASS. ANN. LAWS ch. 258, § 10(d) (Lex-
isNexis 2004); MINN. STAT. ANN. § 3.736, subdiv. 3(c) (West 2005); id. § 466.03, subdiv. 3 (West
employees,3 1 and forty states confer immunity from punitive dam-

ages. 32 Other common immunities conferred on state and local gov-
2001 & Supp. 2006); MISS. CODE ANN. § 11-46-9(1)(i) (West 1999); NEB. REV. STAT. § 13-910(5)
(1997 & Supp. 2005); id. § 81-219(2) (2003 & Supp. 2005); N.J. STAT. ANN. § 59:7-2 (West
1992); N.D. CENT. CODE § 32-12.2-02(3) (h) (1996 repl. & Supp. 2005); OKLA. STAT. ANN. tit. 51,
§ 155(11) (West 2000); OR. REV. STAT. ANN. § 30.265(3)(b) (West 1983 & Supp. 2005); S.C.
CODE ANN. § 15-78-60(11) (2005 & Supp. 2005); TENN. CODE ANN. § 29-20-205(8) (2000 repl.);
TEX. CIv. PRAC. & REM. CODE ANN. § 101.055(1) (Vernon 2005); UTAH CODE ANN. § 63-30d-
301(5)(a) (2004 repl. & Supp. 2005); VT. STAT. ANN. tit. 12, § 5601(e)(2) (2002); VA. CODE
ANN. § 8.01-195.3(5) (2000 repl.); W. VA. CODE § 29-12A-5(a) (8) (LexisNexis 2004 repl.); see also
S.S. Kresge Co. v. Bouchard, 306 A.2d 179, 181 (R.I. 1973) (construing statutory waivers of sov-
ereign immunity not to reach action alleging unlawful overassessment of taxes).
" See ALASKA STAT. § 09.50.250(3) (2004 & Supp. 2005); CONN. GEN. STAT. ANN. § 52-
557n(a) (2) (A) (West 2005); GA. CODE ANN. § 50-21-24(7) (2006); HAW. REV. STAT. § 662-15(4)
(1993 & Supp. 2004); IDAHO CODE ANN. § 6-904(3) (2004); IOWA CODE ANN. § 669.14(4) (West
1998 & Supp. 2006); KY. REV. STAT. ANN. § 44.072 (LexisNexis 1997); MD. CODE ANN., CTs. &
JUD. PROC. § 5-522 (a) (4) (LexisNexis 2002 repl.); MASS. ANN. LAWS ch. 258, § 10(c) (LexisNexis
2004); NEB. REV. STAT. § 13-910(7) (1997 & Supp. 2005); id. § 81-8,219(4) (2003 & Supp. 2005);
N.J. STAT. ANN. § 59:2-10 (West 1992); N.D. CENT. CODE § 32-12.2-02(3) (1996 repl. & Supp.
2005); OHIO REV. CODE. ANN. § 2744.03(A) (2) (LexisNexis 2000 & Supp. 2002); TENN. CODE
ANN. § 9-8-307(d) (1999 rep. & Supp. 2005); id. § 29-20-205(2) (2000 repl.); TEX. CIV. PRAC. &
REM. CODE ANN. § 101.057(2) (Vernon 2005); UTAH CODE ANN. § 63-30d-301 (5) (b) (2004 repl.
& Supp. 2005); VT. STAT. ANN. tit. 12, § 5601 (e) (6) (2002).
"z See ALA. CODE § 6-11-26 (LexisNexis 2005 repl.); ALASKA STAT. § 09.50.280 (2004); ARIz.
REV. STAT. ANN. § 12-820.04 (2003); CAL. GOV'T CODE § 818 (West 1994); COLO. REV. STAT.
§ 24-10-114(4) (2005); FLA. STAT. § 768.28(5) (2005); GA. CODE ANN. § 50-21-30 (2006); HAw.
REV. STAT. § 662-15 (1993 repl. & Supp. 2004); IDAHO CODE ANN. § 6-918 (2004); 745 ILL.
COMP. STAT. ANN. § 10/2-102 (West 2002); IND. CODE ANN. § 34-13-3-4(b) (West 1999 & Supp.
2004); IOWA CODE ANN. §§ 669.4, 670.4(5) (West 1998 & Supp. 2006); KAN. STAT. ANN. § 75-
6105(c) (1997); KY. REV. STAT. ANN. § 65.2002 (LexisNexis 2004 repl.); ME. REV. STAT. ANN. tit.
14, § 8105(5) (2003); MD. CODE ANN., CTS. &JUD. PROC. §§ 5-303(c)(1), -522(a)(1) (LexisNexis
2002 & Supp. 2005); MASS. ANN. LAWS ch. 258, § 2 (LexisNexis 2004); MINN STAT. ANN. § 3.736,
subdiv. 3 (West 2005); id. § 466.04, subdiv. 1(b) (West 2001); MISS. CODE ANN. § 11-46-15(2)
(West 1999); Mo. ANN. STAT. § 537.610(3) (West 2000); MONT. CODE ANN. § 2-9-105 (2005);
NEV. REV. STAT. ANN. § 41.035(1) (LexisNexis 2002 repl.); N.H. REV. STAT. ANN. § 507-B:4(II)
(LexisNexis 1997 repl.); id. § 541-B:14(l) (LexisNexis 2006); N.J. STAT. ANN. § 59:9-2(c) (West
1992 & Supp. 2000); N.M. STAT. ANN. §414-19(C) (LexisNexis 1996); N.D. CENT. CODE §§ 32-
12.1-03(2), -12.2-02(2) (1996 repl. & Supp. 2005); OHIO REV. CODE. ANN. § 2744.05(A) (Lex-
isNexis 2000 & Supp. 2001); OKLA. STAT. ANN. tit. 51, § 154(C) (West 2000 & Supp. 2006); OR.
REV. STAT. ANN. § 30.270(2) (West 1983 & Supp. 1987); 42 PA. CONS. STAT. §§ 8528(c), 8553(c)
(2002); S.C. CODE ANN. § 15-78-120(b) (2005); TENN. CODE ANN. § 9-8-307(d) (1999 repl. &
Supp. 2005); TEX. CV. PRAC. & REM. CODE ANN. § 101.024 (Vernon 2005); VA. CODE ANN.
§ 8.01-195.3 (2000 repl.); W. VA. CODE ANN. § 29-12A-7(a) (LexisNexis 2004 repl.); WIS. STAT.
ANN. § 893.80(3) (West 2006); WvO. STAT. ANN. § 1-39-118(d) (2005); see also Hazen v. Munici-
pality of Anchorage, 718 P.2d 456, 465-66 (Alaska 1986) ("[I]n the absence of statutory au-
thorization punitive damages are not available against a municipality.... ."); City of Hartford v.
Int'l Ass'n of Firefighters, Local 760, 717 A.2d 258, 264 (Conn. App. Ct. 1998) (holding that
punitive damages awards against municipalities are contrary to public policy); Prigge v. Conn.
Dep't of Child & Families, No. X06CV02181467, 2004 Conn. Super. LEXIS 804, at *13 (Conn.
Super. Ct. Mar. 26, 2004) (mem.) (barring punitive damages awards against the state); Schueler
v. Martin, 674 A.2d 882, 889 (Del. Super. Ct. 1996) (finding no legislative intent to allow puni-
tive damages against local governments); Graffv. Motta, 695 A.2d 486, 490 (R.I. 1997) (holding
that there can be no punitive damages awards against a municipality absent a specific waiver of
such immunity).
ernments or their employees include immunity for issuance, denial,

or revocation of a license; 3 a failure to inspect or to make an ade-
quate inspection of property; 4 the adoption or failure to adopt legis-
3 6 acts or omissions in the execu-
lation or other legislative functions;"
tion or enforcement of the law; or
the institution of judicial
" See ALASKA STAT. § 09.65.070(d)(3) (2004); ARIZ. REV. STAT. ANN. § 12-820.02(A) (5)
(2003); CAL. GOV'T CODE § 818.4 (West 1994); CONN. GEN. STAT. ANN. § 52-557n(a) (2)(b)(7)
(West 2005); DEL. CODE. ANN. tit. 10, § 4011 (b)(2) (1999 repl.); GA. CODE ANN. § 50-21-24(9)
(2006); IDAHO CODE ANN. § 6-904B(3) (2004); 745 ILL. COMP. STAT. ANN. § 10/2-104 (West
2002); IND. CODE ANN. § 34-13-3-3(11) (West 1999 & Supp. 2006); IOWA CODE ANN. § 670.4(9)-
(10) (West 1998 & Supp. 2003); KY. REV. STAT. ANN. § 65.2003(3) (c) (LexisNexis 2004 repl.);
ME. REV. STAT. ANN. tit. 14, § 8104-B(2) (2003); id. § 8111(1)(B); MASS. ANN. LAWS ch. 258,
§ 10(e) (LexisNexis 2004); MINN. STAT. ANN. § 3.736, subdiv. 3(k) (West 2005); id. § 466.03,
subdivs. 6d, 10, 15 (Supp. 2006); MISS. CODE ANN. § 11-46-9(1)(h) (West 1999); NEB. REV. STAr.
§ 13-910(4) (1997 & Supp. 2005); id. § 81-8,219(8) (2003 & Supp. 2005); N.J. STAT. ANN. § 59:2-
5 (West 1992); N.D. CENT. CODE §§ 32-12.1-03(3)(f), -12.2-02(3)(d) (1996 repl. & Supp. 2005);
OHIO REV. CODE. ANN. § 2743.01(A), (E) (LexisNexis 2000 repl.); OKLA. STAT. ANN. tit. 51,
§ 155(13) (West 2000); S.C. CODE ANN. § 15-78-60(12) (2005 & Supp. 2005); TENN. CODE ANN.
§ 9-8-307(a) (2) (B) (1999 repl. & Supp. 2005); id. § 29-20-205(3) (2000 repl.); UTAH CODE ANN.
§ 63-30d-301(5) (c) (2004 repl. & Supp. 2005); W. VA. CODE ANN. § 29-12A-5(a) (9) (LexisNexis
2004 repl.); see also R.I. Affiliate, ACLU v. R.I. Lottery Comm'n, 553 F. Supp. 752, 765 n.9
(D.R.I. 1982) (construing Rhode Island's statutory waiver of sovereign immunity not to reach
licensing cases).
" See ALASKA STAT. §09.65.070(d)(1) (2004); ARIZ. REV. STAT. ANN. § 12-820.02(A)(6)
(2003); CAL. GOV'T CODE § 818.6 (West 1995); DEL. CODE ANN. tit. 10, § 4011(b)(1) (1999
repl.); GA. CODE ANN. § 50-21-24(8) (2006); IDAHO CODE ANN. § 6-904B(4) (2004); 745 ILL.
COMp. STAT. ANN. § 10/2-105 (West 2002); IND. CODE ANN. § 34-13-3-3(12) (West 1999 & Supp.
2006); IOWA CODEANN. § 669.14(13) (West 1998 & Supp. 2005); id. § 670.4(6), (12) (West 1998
& Supp. 2003); KAN. STAT. ANN. § 75-6104(k) (1997 & Supp. 2005); Ky. REV. STAT. ANN.
§ 65.2003(e) (LexisNexis 2004 repl.); MASS. ANN. LAWS ch. 258, § 10(f) (LexisNexis 2004); NEB.
REV. STAT. § 13-910(3) (1997 & Supp. 2005); id. § 81-8,219(7) (2003 & Supp. 2005); NEV. REV.
STAT. ANN. § 41.033 (LexisNexis 2002 repl.); N.J. STAT. ANN. § 59:2-6 (West 1992); N.Y. GEN.
MUN. LAW § 205-d (McKinney 1999); N.D. CENT. CODE §§ 32-12.1-02(3) (f), -12.2-02(3) (f) (1996
repl. & Supp. 2005); OHIO REV. CODE ANN. § 2743.01 (A), (E) (LexisNexis 2000 & Supp. 2003);
OKLA. STAT. ANN. tit. 51, § 155 (West 2000); S.C. CODE ANN. § 15-78-60(13) (2005 & Supp.
2005); TENN. CODE ANN. § 9-8-307(a) (2) (A) (1999 repl. & Supp. 2005); id. § 29-20-205(4) (2000
repl.); UTAH CODE ANN. § 63-30d-301 (5) (d) (2004 repl. & Supp. 2005); W. VA. CODE ANN. § 29-
12A-5(a) (10) (LexisNexis 2004 repl.); Wis. STAT. ANN. § 893.80(lp) (West 2006).
" See CAL. GOV'T CODE § 818.2 (West 1995); GA. CODE ANN. § 50-21-24(4) (2006); 745 ILL.
COMP. STAT. ANN. § 10/2-103 (West 2002); IND. CODE ANN. § 34-13-3-3(8) (West 1999 & Supp.
2006); KAN. STAT. ANN. § 75-6104(a) (1997 & Supp. 2005); KY. REV. STAT. ANN. § 65.2003(3) (a)
(LexisNexis 2004 repl.); ME. REV. STAT. ANN. tit. 14, §§ 8104-B(1), 8111(1)(A) (2003); Miss.
CODE ANN. § 11-46-9(1)(a), (e) (West 1999); MONT. CODE ANN. § 2-9-111(2) (2005); N.J. STAT.
ANN. § 59:2-3(b) (West 1992); N.D. CENT. CODE §§ 32-12.1-03(3) (b), -12.2-02(3) (c) (1996 repl.
& Supp. 2005); OHIO REV. CODE ANN. § 2744.03(A)(1) (LexisNexis 2000 repl. & Supp. 2002);
OKLA. STAT. ANN. tit. 51, § 155(1) (West 2000); S.C. CODE ANN. § 15-78-60(1), (2), (4) (2005 &
Supp. 2005); VA. CODE ANN. § 8.01-195.3(2) (2000); W. VA. CODE ANN. § 29-12A-5(a)(4) (Lex-
isNexis 2004 repl.).
See CAL. GOV'T CODE § 821.4 (West 1995); CONN. GEN. STAT. ANN. § 52-557n(a)(2)(b)(8),
(9) (West 2005); IDAHO CODE ANN. § 6-904B(1) (2004); 745 ILL. COMP. STAT. ANN. § 10/2-202
(West 2002); IOWA CODE ANN. § 669.14(3), (11) (West 1998 & Supp. 2006); KAN. STAT. ANN.
§ 75-6104(c) (1997 & Supp. 2005); KY. REV. STAT. ANN. § 44.073(13)(d) (LexisNexis 1997 repl.
& Supp. 2005); id. § 65.2003(3)(b) (LexisNexis 2004 repl.); MIsS. CODE ANN. § 11-46-9(1)(j)
JOURNAL OF CONSTTUTFIONAL LA W [Vol. 9:3
administrative proceedings; 31 the plan or design for public improve-

ments;' the condition of property or facilities 4 used for recreational
39
or of unimproved
purposes police public property;4° a failure to provide
adequate service or protection or to provide adequate jails or
(West 1999); N.J. STAT. ANN. § 59:2-4 (West 1992); OKLA. STAT. ANN. tit. 51, § 155(4) (West
2000); S.C. CODE ANN. § 15-78-60(4) (2005 & Supp. 2005); W. VA. CODE ANN. § 29-12A-5 (a) (4)
(LexisNexis 2004 repl.); see also Dang v. Ehredt, 977 P.2d 29, 35-36 (Wash. Ct. App. 1999)
(holding that law-enforcement officers possess common-law immunity for reasonable enforce-
ment decisions).
" See CAL. GOV'T CODE § 821.6 (West 1995); CONN. GEN. STAT. ANN. § 52-557n (a) (2) (b) (5)
(West 2005); GA. CODE ANN. § 50-21-24(5) (2006); 745 ILL. COMP. STAT. ANN. § 10/2-208 (West
2002); IND. CODE ANN. § 34-13-3-3(6) (West 1999 & Supp. 2006); ME. REV. STAT. ANN. tit. 14,
§§ 8104-B(4), 8111(1)(D) (2003); N.J. STAT. ANN. § 59:2-3(b) (West 1992); OHIO REV. CODE
ANN. § 2744.03(A)(1) (LexisNexis 2000 repl. & Supp. 2002); OKLA. STAT. ANN. tit. 51, § 155(2)
(West 2000); S.C. CODE ANN. § 15-78-60(23) (2005 & Supp. 2005); TENN. CODE ANN. § 29-20-
205(5) (2000 repl.); UTAH CODE ANN. § 63-30d-301 (5) (e) (2004 repl. & Supp. 2005); VA. CODE
ANN. § 8.01-195.3(6) (2000); W. VA. CODE ANN. § 29-12A-5(a)(2) (LexisNexis 2004 repl.).
See CAL. GOV'T CODE § 830.6 (West 1995); IDAHO CODE ANN. § 6-904(7) (2004); 745 ILL.
COMP. STAT. ANN. § 10/3-103 (West 2002); IOWA CODE ANN. §§ 669.14(8), 670.4(7)-(8) (West
1998 & Supp. 2006); KAN. STAT. ANN. § 75-6104(m) (1997 & Supp. 2005); ME. REV. STAT. ANN.
tit. 14, § 8104-A(2)(A), (4) (2003); MINN. STAT. ANN. § 466.03, subdivs. 18, 20 (West 2001 &
Supp. 2006); MISS. CODE ANN. § 1146-9(1) (p) (West 1999); NEB. REV. STAT. § 13-910(11) (1997
& Supp. 2005); id. §81-8,219(11) (2003 & Supp. 2005); N.J. STAT. ANN. § 59:4-6 (West 1992);
TEX. Ctv. PRAC. & REM. CODE ANN. §§ 101.063, 101.064(b) (Vernon 2005); UTAH CODE ANN.
§ 63-30d-301 (5) (n) (2004 repl. & Supp. 2005); VT. STAT. ANN. tit. 12, § 5601 (e) (8) (2002); WYO.
STAT. ANN. § 1-39-120 (2005).
39 See CAL. GOV'T CODE §§ 831.21, .4 (West 1995); CONN. GEN. STAT. ANN. § 52-
557n(a) (2) (b) (4) (West 2005); HAw. REV. STAT. §§ 662-15(7), -19 (1993 repl. & Supp. 2004);
745 ILL. COMP. STAT. ANN. § 10/3-106 (West 2002); IND. CODE ANN. § 34-13-3-3(4), (5) (West
1999 & Supp. 2006); IOWA CODE ANN. § 670.4(15) (West 1998 & Supp. 2006); KAN. STAT. ANN.
§ 75-6104(o) (1997 & Supp. 2005); ME. REV. STAT. ANN. tit. 14, § 8104-A(2) (A) (2003); MINN.
STAT. ANN. § 3.736, subdiv. 3(i) (West 2005); id. § 466.03, subdivs. 6e, 6f, 15 (Supp. 2006); N.H.
REV. STAT. ANN. § 507-B:11 (LexisNexis 1997); N.J. STAT. ANN. § 59:2-7 (West 1992); S.C. CODE
ANN. § 15-78-60(16), (26) (2005 & Supp. 2005).
40 See CONN. GEN. STAT. ANN. § 52-557n(a) (2) (b) (1) (West 2005); IND. CODE ANN. § 34-13-3-
4
3(1) (West 1999 & Supp. 2006); KAN. STAT. ANN. § 75-610 (p) (1997 & Supp. 2005); ME. REV.
STAT. ANN. tit. 14, § 8104-A(2) (A) (2003); MINN STAT. ANN. § 3.736, subdiv. 3(g) (West 2005);
id. § 466.03, subdivs. 13, 15 (Supp. 2006); N.J. STAT. ANN. § 59:4-8, -9 (West 1992); N.D. CENT.
CODE § 32-12.2-02(3)(k) (1996 repl. & Supp. 2005); OKLA. STAT. ANN. tit. 51, § 155(9)-(10)
(West 2000); S.C. CODE ANN. § 15-78-60(10) (2005 & Supp. 2005); UTAH CODE ANN. § 63-30d-
301(5)(k) (2004 repl. & Supp. 2005); W. VA. CODE ANN. § 29-12A-5(a)(7) (LexisNexis 2004
repl.).
" SeeARiZ. REV. STAT. ANN. § 12-820.02(A) (1), (7)-(8) (2003); CAL. GOV'T CODE §§ 845,846
(West 1995); GA. CODE ANN. § 50-21-24(6) (2006); 745 ILL. COMP. STAT. ANN. § 10/4-102, -107
(West 2002); KAN. STAT. ANN. § 75-6104(n) (1997 & Supp. 2005); Ky. REV. STAT. ANN.
§ 44.073(13) (e) (LexisNexis 1997 repl. & Supp. 2005); MASS. ANN. LAWS ch. 258, § 10(h) (Lex-
isNexis 2004); MISS. CODE ANN. § 1146-9(1)(c) (West 1999); NJ. STAT. ANN. § 59:5-4 (West
1992); N.D. CENT. CODE §§ 32-12.1-03(3)(f), -12.2-02(3)(f) (1996 repl. & Supp. 2005); OHIO
REV. CODE ANN. § 2743.02 (A) (3) (a) (LexisNexis 2000 repl. & Supp. 2005); OKLA. STAT. ANN.
tit. 51, § 155 (West 2000); TEX. CIV. PRAc. & REM. CODE ANN. § 101.055(3) (Vernon 2005); W.
VA. CODE ANN. § 29-12A-5(a)(5) (LexisNexis 2004 repl.); Wis. STAT. ANN. § 893.80(6) (West
2006).
Feb. 2007] A THEORY OF GOVERNMENTAL DAMAGES LIABILTY 809
other corrections or penal facilities; 42 the probation, parole, release,

or escape of arrestees, convicts, or prisoners; 43 a failure to provide
adequate firefighting or other emergency service 44 a failure to pro-
vide adequate medical care a4 or to prevent disease or impose a quar-
antine; 4' and specified unintentional torts.47 Some immunity statutes,
rather than granting categorical immunities, delineate the circum-
stances under which governmental defendants can be held liable-
usually involving injuries caused by the condition or use of public
12 See ARIZ. REV. STAT. ANN. § 12-820,02(A)(4) (2003); CAL. GOV'T CODE § 845.2 (West
1995); 745 ILL. COMP. STAT. ANN. § 10/4-103 (West 2002); IOWA CODE ANN. § 669.14(6) (1998
& Supp. 2006); MINN. STAT. ANN. § 3.736, subdiv. 3(m) (West 2005); id. § 466.03, subdiv. 12
(Supp. 2006); MISS. CODE ANN. § 11-46-9(1)(d), (in) (West 1999); MONT. CODE ANN. § 2-9-
108(2) (2005); N.J. STAT. ANN. § 59:5-1 (West 1992); OKLA. STAT. ANN. tit. 51, § 155(24) (West
2000); S.C. CODE ANN. § 15-78-60(25) (2005 & Supp. 2005); S.D. CODIFIED LAWS § 3-21-8
(2004); UTAH CODE ANN. § 63-30d-301(5) (j) (2004 repl. & Supp. 2005); W. VA. CODE ANN. § 29-
12A-5 (a) (14) (LexisNexis 2004 repl.).
" SeeARiZ. REV. STAT. § 12-820.02(A) (2)-(3) (2003); CAL. GOV'T CODE § 845.8 (West 1995);
IDAHO CODE ANN. § 6-904A(2) (2004); 745 ILL. COMP. STAT. ANN. § 10/4-106 (West 2002); IND.
CODE ANN. § 34-13-3-3(17) (West 1999 & Supp. 2006); KAN. STAT. ANN. § 75-6102(d)(1) (1997
& Supp. 2005); MAss. ANN. LAws ch. 258, § 10(i) (LexisNexis 2004); N.D. CENT. CODE §§ 32-
12.1-03(3)(f), -12.2-02(3)(f) (1996 repl. & Supp. 2005); N.J. STAT. ANN. § 59:5-2 (West 1992);
OKLA. STAT. ANN. tit. 51, § 155(22), (24) & (25) (West 2000); S.C. CODE ANN. § 15-78-60(21)
(2005 & Supp. 2005); S.D. CODIFIED LAWS § 3-21-9 (2004); TENN. CODE ANN. § 9-8-307(a) (2)
(1999 repl. & Supp. 2005); VT. STAT. ANN. tit. 24, § 296a (2005); W. VA. CODE ANN. § 29-12A-
5(a) (13) (LexisNexis 2004 repl.).
" See ALASKA STAT. § 09.65.070(c) (2004); CAL. GOV'T CODE §§ 850-850.4 (West 1995); D.C.
CODE ANN. § 2412 (LexisNexis 2001); 745 ILL. COMP. STAT. ANN. §§ 10/5-101, -103 (West
2002); IOWA CODE ANN. § 670.4(11) (West 1998); KAN. STAT. ANN. § 75-6104(j) (1997 repl. &
Supp. 2005); LA. REV. STAT. ANN. § 9:2793.1 (1997); MD. CODE ANN., CTS. &JUD. PROC. § 5-604
(LexisNexis 2002 repl.); MASS. ANN. LAWS ch. 258, § 10(g) (LexisNexis 2004); MISS. CODE ANN.
§ 11-46-9(1)(c) (West 1999); N.Y. GEN. MUN. LAW § 205-b (McKinney 1999); N.D. CENT. CODE
§ 32-12.1-03(3) (f), -12.2-02(3)(f) (1996 repl. & Supp. 2005); OHIO REV. CODE ANN. § 9.60 (Lex-
isNexis 2001 repl. & Supp. 2003); id. § 2743.01 (A), (E) (LexisNexis 2000 repl. & Stipp. 2002);
id. § 3737.221 (LexisNexis 2005 repl.); S.C. CODE ANN. § 23-47-70 (2005); S.D. CODIFIED LAws
§ 33-15-38 (2004); TEX. CIV. PRAC. & REM. CODE ANN. § 101.055(2), (3) (Vernon 2005); UTAH
CODE ANN. § 63-30d-301(5)(o), (p) (2004 repl. & Supp. 2005); W. VA. CODE ANN. § 29-12A-
5(a) (5) (LexisNexis 2004 repl.).
" See CAL. GOV'T CODE §§ 855.4, .6 (West 1995); 745 ILL. COMP. STAT. §§ 10/6-105, -106
(West 2002); MINN. STAT. ANN. § 3.736, subdiv. 3(1) (West 2005); id. § 466.03, subdiv. 11 (West
2001 & Supp. 2002); MISS. CODE ANN. § 11-46-9(1) (r) (West 1999); N.J. STAT. ANN. §§ 59:6-2, -4
to -6 (West 1992).
" See ALASKA STAT. § 09.50.250(1) (2004 & Supp. 2005); CAL. GOV'T CODE § 855.4 (West
1995); IDAHO CODE ANN. § 6-904(2) (2004); 745 ILL. COMP. STAT. ANN. § 10/6-104 (West 2002);
IOWA CODE ANN. § 669.14(3) (West 1998 & Supp. 2006); MISS. CODE ANN. § 11-46-9(1)(k)
(West 1999); NEB. REV. STAT. § 13-910(6) (1997 & Supp. 2005); id. § 81-8,219(3) (2003 & Supp.
2005); N.H. REV. STAT. ANN. § 541-B:21(I) (2006); N.J. STAT. ANN. § 59:6-3 (West 1992); S.C.
CODE ANN. § 15-78-60(18) (2005 & Supp. 2005); UTAH CODE ANN. § 63-30d-201(2) (2004 repl.);
VT. STAT. ANN. tit, 12, § 5601 (e) (3) (2002).
,7 See IND. CODE ANN.§ 34-13-3-3(14) (West 1999 & Supp. 2006); ME. REV. STAT. ANN. tit. 14,
§8111(1)(E) (2003); NEV. REV. STAT. ANN. § 41.0334 (2002); N.H. REV. STAT. ANN. § 541-
B:19(d) (2006); OKLA. STAT. ANN. tit. 51, § 155(17) (West 2000); 42 PA. CONS. STAT. § 8546(2)
(2002); W. VA. CODE ANN. § 29-12A-5(a) (12) (LexisNexis 2004 repl.).
JOURNAL OFCONSTFiTUTIONAL LAW [Vol. 9:3
property-and otherwise grant immunity. 8 In addition, forty-two

states limit the damages recoverable from a governmental defendant
or a public employee. 9
" See ALA. CODE § 11-47-190 (LexisNexis 1992 repl. & Supp. 1994) (allowing municipalities
to be liable for torts of employees within the scope of employment and for conditions on
streets, alleys, and public ways); COLO. REV. STAT. § 24-10-106 (2005) (allowing public entities to
be liable for negligent acts arising from the operation of nonemergency vehicles and the main-
tenance of hospitals, jails, correctional facilities, public buildings, and roadways); DEL. CODE
ANN. tit. 10, §§ 4011 (c), 4012 (1999 repl.) (allowing municipal employees to be liable for "wan-
ton negligence or willful and malicious intent"; allowing municipalities to be liable for negli-
gent acts arising from the ownership, use, or maintenance of vehicles, equipment, or aircraft;
construction, operation, or maintenance of buildings; and the sudden and accidental discharge
of pollutants); D.C. CODE ANN. § 2-412 (LexisNexis 2005 repl.) (holding the district liable for
damages caused by the operation of vehicles); ME. REV. STAT. ANN. tit. 14, §§ 8104-A(1)(A)-
(G) (2003) (allowing liability for negligent acts or omissions in the ownership, maintenance, or
use of any vehicle; negligent construction, maintenance, or operation of a public building; neg-
ligent and sudden or accidental discharge of pollutants; and negligent construction, cleaning,
or repair of roads); MICH. COMP. LAWS ANN. §§ 691.1402(1), .1405, .1407, .1413, .1417(3) (West
2000 & Supp. 2006) (preserving state and local government immunity except for proprietary
functions, negligent operation of motor vehicles, negligent medical care except for hospitals
owned by department of community health or department of corrections, overflow of sewer
drains caused by a negligent failure to repair, and negligent maintenance of highways); Mo.
ANN. STAT. §§ 537.600, .610.1 (West 2000 & Supp. 2005) (allowing state and local governments
to be liable for operation of motor vehicles and dangerous conditions on public property);
N.M. STAT. ANN. §§ 41-4-5 to -12 (LexisNexis 1996 repl.) (allowing government entities to be
liable for negligence in the operation of vehicles, negligence in the operation or maintenance
of public property or equipment, and battery, false imprisonment, and other intentional torts);
N.C. GEN. STAT. §§ 115C-42, 153A-435, 160A-485 (2005) (immunizing local governments and
employees only to the extent they purchase insurance); 42 PA. CONS. STAT. § 8522 (2002) (al-
lowing the State and state employees to be liable for negligence in operation of vehicles, medi-
cal malpractice, and other situations); TENN. CODE ANN. §§ 29-20-201, -202 (2000 repl.) (allow-
ing local governments to be liable for willful or wanton conduct, gross negligence, the
operation of vehicles, and unsafe conditions on public property); UTAH CODE ANN. § 63-30d-
301 (2004 repl. & Supp. 2005) (waiving immunity for damage to property held for purposes of
forfeiture, defective or dangerous conditions of public property except when latent, and negli-
gent acts of public employees); WIS. STAT. ANN. §§ 345.05, 893.81, .83 (West 2006) (allowing
local-governmental liability for riots, failure to repair highways, and negligent operation of a
motor vehicle); WvO. STAT. ANN. §§ 1-39-104 to -108 (2005) (establishing liability for negligent
operation of vehicles, negligent care of buildings and recreation areas, and tortious conduct of
police officers); see also Scott v. Savers Prop. & Cas. Ins. Co., 663 N.W.2d 715, 721 (Wis. 2003)
(permitting only actions against public officers involving "(1) ministerial duties .... (2) duties
to address a known danger, (3) actions involving professional discretion, and (4) actions that
are malicious, willful, and intentional").
" See ALA. CODE § 11-93-2 (LexisNexis 1994 repl.) (limiting damages judgments against
government entities to $100,000 per claimant or $300,000 per occurrence); id. §§ 41-9-62, -68,
-70 (LexisNexis 2000 repl.) (permitting claims against the State heard by the Board of Adjust-
ment subject to worker's compensation caps); ALASKA STAT. § 09.50.270 (2004) (permitting the
Department of Administration to approve payment or recommend an appropriation to the leg-
islature); ARK. CODE ANN. §§ 19-10-204, -215 (1998 repl. & Supp. 2003) (allowing claims against
the State to be heard by the State Claims Commission, which can award up to $10,000); COLO.
REV. STAT. § 24-10-114 (2005) (limiting damages awards against the State to $150,000 per per-
son and $600,000 per incident); CONN. GEN. STAT. §§ 4-142, -159 (1998) (permitting claims
against the State to be heard by the Office of Claims Commissioner, which can award up to
$7,500 or recommend a greater award to the legislature); DEL. CODE ANN. tit. 10, § 4013 (1999
repl.) (limiting damages judgments against the State to $300,000 per occurrence); FLA. STAT.
§ 768.28(5), (8) (2005) (limiting damages judgments against the State to $100,000 per claimant
and $200,000 per occurrence); GA. CODE ANN. § 50-21-29(b) (2006) (limiting damages judg-
ments against the State to $1 million per plaintiff and $3 million per occurrence); IDAHO CODE
ANN. §§ 6-924, -926 (2004) (limiting damages judgments against state entities to $500,000 per
occurrence); 705 ILL. COMP. STAT. ANN. § 505/8 (West 1999 & Supp. 2003) (permitting Court
of Claims to enter judgments against the State up to $100,000 per claimant); IND. CODE ANN.
§ 34-13-3-4(a) (West 1999 & Supp. 2006) (limiting damages judgments against the State to
$300,000 per person for actions accruing before 2006); KAN. STAT. ANN. § 75-6105(a) (1997)
(limiting damages judgments against the State to $500,000 per incident); Ky. REV. STAT. ANN.
§§ 44.070, .072 (LexisNexis 1997 repl. & Supp. 2005) (permitting claims against the State to be
heard by the Board of Claims, which can award $200,000 per person and $350,000 per inci-
dent); LA. REV. STAT. ANN. § 13:5106(B) (2006) (limiting damages judgments against the State
to $500,000 per person); ME. REV. STAT. ANN. tit. 14, §§ 8104-D, 8105(1) (2003) (limiting dam-
ages judgments against public employees to $10,000 per occurrence and against the State to
$400,000 per occurrence); MD. CODE ANN., Crs. & JUD. PROC. § 5-303(a)(1) (LexisNexis 2002
repl. & Supp. 2005) (limiting damages judgments against local governments to $200,000 per
person and $500,000 per incident); MD. CODE ANN., STATE GOVT § 12-104(a)(2) (LexisNexis
2004 repl.) (limiting damages judgments against the State to $200,000 per person); MAss. ANN.
LAWS ch. 258, § 2 (LexisNexis 2004) (capping negligence liability of public employees at
$100,000); MICH. COMP. LAws § 600.6419(1) (2000) (permitting the Court of Claims to award
up to $1,000 against the State); MINN. STAT. ANN. § 3.736, subdiv. 4 (West 2005) (limiting dam-
ages judgments against the State to $1 million per occurrence and $300,000 for wrongful
death); id. § 466.04, subdiv. 1(a) (West 2001 & Supp. 2002) (limiting damages judgments
against municipalities to $1 million per occurrence and $300,000 for wrongful death); Miss.
CODE ANN. § 11-46-15(2)(c) (West 1999) (limiting damages judgments against government en-
tities to $500,000 per occurrence); MO. ANN. STAT. § 537.610.1(2) (West 2000) (limiting dam-
ages judgments against the State to $300,000 per plaintiff and $2 million per occurrence);
MONT. CODE ANN. § 2-9-108(1) (2005) (limiting damages judgments against the State to
$750,000 per claim and $1.5 million per occurrence); NEB. REV. STAT. §§ 13-915, -922 (1997 &
Supp. 2005) (limiting damages judgments against local governments and employees to $1 mil-
lion per person and $5 million per incident); id. §§ 81-8,209, -8,211, -8,214, -8,215 (2003 &
Supp. 2005) (permitting the State Claims Board to award damages against the State up to
$5,000, or $25,000 if unanimously approved); NEV. REV. STAT. ANN. § 41.035 (LexisNexis 2002
repl.) (limiting damages judgments against the State to $50,000); N.H. REV. STAT. ANN. § 507-
B:4 (LexisNexis 1997 repl.) (limiting damages judgments against local governments and em-
ployees to $150,000 per claimant and $500,000 per incident); id. § 541-B:14 (LexisNexis 2006
repl.) (limiting damages judgments against the State and state employees to $250,000 per claim-
ant and $2 million per incident); N.M. STAT. ANN. § 41-4-19(A) (LexisNexis 1996 repl. & Supp.
2004) (limiting damages judgments against the State to $100,000 per occurrence for property
damage, $300,000 per occurrence for medical expenses, and $400,000 per person or $750,000
per occurrence for all other damages); N.Y. CT. CL. ACT § 9 (McKinney 2005) (capping claims
against state and local governments heard by the Court of Claims); N.C. GEN. STAT. §§ 143-291,
-299.1, -299.2, -299.4, -300.1, -300.6, -3001A (2005) (capping damages judgments against the
State, agencies, and employees at $150,000 per person and $500,000 per occurrence); N.D.
CENT. CODE §§ 32-12.1-03(3) (2), -12.2-02(2) (1996 repl. & Supp. 2005) (limiting damages
judgments against political subdivisions to $250,000 per person and $500,000 per occurrence
and against the State to $250,000 per person and $1 million per occurrence); OHIO REV. CODE
ANN. §§ 2743.02(D), 2744.05(B) (LexisNexis 2000 repl. & Supp. 2002) (permitting the State to
be sued in the Court of Claims with reduction of recovery for collateral recoveries by plaintiff
and limiting noneconomic damages judgments against local governments to $250,000); OKLA.
STAT. ANN. tit. 51, § 154 (West 2000) (capping damages awards at $25,000 per person for prop-
erty damage, $125,000 for other losses, $175,000 in cities with a population of 300,000 or more,
$200,000 for medical negligence, and $175,000 for a wrongful felony conviction); OR. REV.
STAT. ANN. § 30.270 (West 1983 & Supp. 1987) (limiting damages to $50,000 per person for
JOURNAL OF CONSTITIIONAL LA W [Vol. 9:3
State statutes usually grant public employees additional protec-

tions from the threat of liability. ° Public employers are usually re-
quired by statute to indemnify their employees or otherwise pay
judgments against those employees arising from torts committed
within the scope of their employment, although indemnity is gener-
ally not required in cases of egregious individual misconduct. 5 Like
property damage, $100,000 for special damages, $100,000 for general damages for personal in-
juries, and $500,000 in the aggregate arising from a single incident); 42 PA. CONS. STAT.
§§ 8549, 8528(b), 8553 (2002) (limiting damages judgments against the State and state employ-
ees to $250,000 per person and $1 million aggregate and against local governments and em-
ployees to $500,000 total); R.I. GEN. LAWS §§ 9-31-2, -3, 24-5-13(b) (1997) (limiting damages
judgments against the State to $100,000 except for proprietary functions and capping local-
governmental liability for potholes at $300); S.C. CODE ANN. § 15-78-190 (2005) (limiting dam-
ages judgments against government entities to $300,000 per person and $600,000 per occur-
rence); S.D. CODIFIED LAWS §§ 3-22-8, -10, 21-32-16, -19, 21-32A-1 (2004) (permitting the State
to be liable up to $2,000 and only beyond to the extent the defendant is covered by insurance);
TENN. CODE ANN. §§ 9-8-108(a)(7) (A), -307(e) (1999 repl. & Supp. 2005) (referring specified
claims against the State to the Claims Commission with a cap of $300,000 per claimant and $1
million per occurrence and other claims against the State to Board of Claims with $1 million
aggregate cap); TEX. Civ. PRAC. & REM. CODE ANN. § 101.023(a)-(c), 108.002 (Vernon 2005)
(limiting state liability to $250,000 per person and $500,000 per occurrence; nonmunicipal-
local-governmental liability to $100,000 per person and $300,000 per occurrence for personal
injury and $100,000 per occurrence for property damage; municipal liability to $250,000 per
person and $500,000 per occurrence for personal injury and $100,000 per occurrence for
property damage; and public-employee liability to $100,000 for personal injury, death, or depri-
vation of a right, privilege, or immunity, and $100,000 for property damage); UTAH CODE ANN.
§ 63-30d-604(1) (2004 repl. & Supp. 2006) (capping personal injury at $553,500 per person and
$1,107,000 per occurrence and property damage at $221,400 per occurrence); VA. CODE ANN.
§ 8.01-195.3 (2000) (capping damages against the State at $100,000 or available insurance); W.
VA. CODE §§ 14-2-2, -13, 29-12-1, -12A-7(b) (LexisNexis 2004 repl.) (directing claims against the
State to the Court of Claims, which can issue judgments within appropriated limits, requiring
insurance in specified cases, and capping noneconomic damages at $500,000 per person); WIs.
STAT. ANN. § 893.80(3) (West 2006) (limiting damages judgments against governmental subdi-
visions to $50,000 per person); WYO. STAT. ANN. § 1-39-118(a) (2005) (limiting damages judg-
ments against government entities to $250,000 per plaintiff and $500,000 per incident).
In some states, plaintiffs can circumvent sovereign immunity by suing an individual public
employee in tort when the claim is deemed to be based on the employee's abuse of authority
through his own wrongful conduct. See, e.g., Curry v. Woodstock Slag Corp., 6 So. 2d 479, 480
(Ala. 1942) (noting that sovereign "immunity does not extend to officers" who "act[] not within
their authority... injurious[ly] to the rights of others"); Currie v. Lao, 592 N.E.2d 977, 983 (Il.
1992) (holding that a state trooper could be sued where the duty he breached did not arise out
of his status as a government employee); Guffey v. Cann, 766 S.W.2d 55, 58 (Ky. 1989) (explain-
ing that under Kentucky law "[t]he doctrine of sovereign immunity [does not] protect state
employees[] sued in their individual capacities"); Morell v. Balasubramanian, 514 N.E.2d 1101,
1103 (N.Y. 1987) (holding that the New York Court of Claims Act did not abolish the common-
law practice of holding state employees individually liable for tortious acts); see also Coffee
County Sch. Dist. v. Snipes, 454 S.E.2d 149, 151 (Ga. Ct. App. 1995) (noting that state employ-
ees lack "immunity for ministerial acts negligently performed or for ministerial or discretionary
acts performed with malice").
" See ALA. CODE § 11-47-24 (LexisNexis 1992 repl.); ARIZ REV. STAT. ANN. §§ 12-820.04, 41-
621(P) (2003); ARK. CODE ANN. § 21-9-203(a) (2006); CAL. GOV'T CODE § 825 (West 1995);
COLO. REV. STAT. § 24-10-118(2) (a) (2005); CONN. GEN. STAT. ANN. §§ 5-141d(a), 7-465 (West
1998 & Supp. 2006); DEL. CODE ANN. tit. 10, § 4002 (1999 repl.); D.C. CODE ANN. § 1-109(a)
Feb.2007] A THEORY OF GOVERNMENTAL DAMAGES LIABILITY 813
the FTCA, some states bar actions against public employees based on
tortious
5
conduct occurring within the scope of their public employ-
ment.
3. Asymmetry with Private Tort Liability

We have seen that federal, state, and local governments and their
employees are free to engage in a broad swath of tortious activities
(LexisNexis 2003 repl.); GA. CODE ANN. § 45-9-22(a) (2002); IDAHO CODE ANN. § 6-903(b)-(c)
(2005); 745 ILL. COMP. STAT. ANN. 10/9-102 (West 2002); IND. CODE ANN. §§ 34-13-3-5, -4-1
(West 1999 & Supp. 2006); IOWA CODE ANN. §§ 669.21, .22, 670.8 (West 1998); KAN. STAT. ANN.
§ 75-6109 (1997); KY. REV. STAT. ANN. § 65.2005 (LexisNexis 2004 repl.); LA. REV. STAT. ANN.
§ 13:5108.1 (2006); ME. REV. STAT. ANN. tit. 14, § 8112 (2003); MD. CODE ANN., CTS. &JUD.
PROC. § 5-303 (LexisNexis 2002 & Supp. 2005); MD. CODE ANN., STATE GOV'T § 12404 (Lex-
isNexis 2004 repl.); MASS. ANN. LAWS ch. 258, §§ 9, 9A, 13 (LexisNexis 2004); MICH. COMP.
LAWS ANN. § 466.07 (West 2001); id. § 691.1408(1) (West 2000); MINN STAY. ANN. § 3.736, sub-
div. 9 (West 2005); id. § 466.07 (West 2001 & Supp. 2002); MISS. CODEANN. §§ 11-46-7(3) (West
1999); MO. ANN. STAT. § 105.711(2)(2) (West 2000 & Supp. 2005); MONT. CODE ANN. § 2-9-305
(2005); NEB. REV. STAT. § 13-1801 (1997); id. § 81-8,239.05 (2003); NEV. REV. STAT. ANN.
§ 41.0349 (LexisNexis 2002 repl.); N.H. REV. STAT. ANN. §§ 29-A:2, 31:105 (2001 repl. & Supp.
2003); id. § 99-D:2 (2004 repl.); N.J. STAT. ANN. § 59:10-1, -4 (West 1992); N.M. STAT. ANN. § 41-
4-4(C) to (E) (LexisNexis 1996 repl.); N.Y. PUB. OFF. LAW §§ 17(2), (3), 18 (McKinney 2001);
N.C. GEN. STAT. § 143-300.3 (2005); N.D. CENT. CODE §§ 32-12.1-04(4), -12.2-03(4) (1996 repl.
& Supp. 2005); OHIO REV. CODE ANN. § 9.87 (LexisNexis 2001 repl. & Supp. 2003); id.
§ 2744.07(A) (LexisNexis 2000 repl. & Supp. 2001); OKLA. STAT. ANN. tit. 51, § 162 (West
2000); OR. REV. STAT. ANN. § 30.285 (West 1983 & Supp. 1987); 42 PA. CONS. STAT. § 8550
(2002); R.I. GEN. LAWS § 9-31-12(a) (1997); id. § 45-15-16 (1999); S.C. CODE ANN. § 1-11-440
(2005); id. § 12-4-325(A) (2000); S.D. CODIFIED LAWS §§ 3-19-1 to -3 (2004); TENN. CODE ANN.
§ 9-8-112(a) (1999 repl.); id. § 29-20-310 (2000 repl. & Supp. 2001); TEx. CIV. PRAC. & REM.
CODE ANN. §§ 102.002, 104.001 (Vernon 2005); UTAH CODE ANN. § 63-30d-603(1)(b) (2004
repl.); VT. STAT. ANN. tit. 12, § 5601 (2002); id. tit. 24, § 901a(b) (2005); VA. CODE ANN. § 2.2-
1837 (2005 repl.); id. § 15.2-1520 (2000); WASH. REV. CODE ANN. §§ 4.92.075, .96.075 (West
2006); W. VA. CODE ANN. § 29-12A-11 (a) (2) (LexisNexis 2004 repl.); WIS. STAT. ANN. § 895.46
(West 2006); WVO. STAT. ANN. § 1-39-104(c) (2005); Gamble v. Fla. Dep't of Health & Rehabili-
tative Servs., 779 F.2d 1509, 1517-18 (11th Cir. 1986) (construing Florida law to require the
indemnification of state officers absent intentional misconduct); Livesay v. Baltimore, 862 A.2d
33, 38 (Md. 2004) (construing Maryland law to require the indemnification of local employees).
5 See ALASKA STAT. § 09.50.253 (2004); ARK. CODE ANN. § 19-10-305(a)
(2005) (state em-
ployees); FLA. STAT. § 768.28(9)(a) (2005); GA. CODE ANN. § 50-21-25(a) (2006); HAW. REV.
STAT. § 662-10 (1993 repl.); IND. CODE § 34-13-3-5(c) (West 1999 & Supp. 2005); MD. CODE
ANN., CTS. &JUD. PROC. § 5-522(b) (LexisNexis 2002 repl.); MASS. ANN. LAWS ch. 258, § 2 (Lex-
isNexis 2004); MICH. COMP. LAWS ANN. § 691.1407(2) (West 2000); MISS. CODE ANN. § 11-46-
7(2) (West 1999); MONT. CODE ANN. § 2-9-305(5)-(6) (2005); OHIO REV. CODE. ANN. § 9.86
(LexisNexis 2001 repl.); OR. REV. STAT. ANN. § 30.265(1) (West 1983 & Supp. 2005); R.I. GEN.
LAWS § 9-31-12(b) (1997); S.C. CODE ANN. § 15-78-70 (2005 & Supp. 2005); S.D. CODIFIED LAWS
§ 21-32-17 (2004); TENN. CODE ANN. § 9-8-307(h) (1999 repl.); id. § 29-20-310(b) (2000 repl.);
UTAH CODE ANN. § 63-30d-202(3) (2004 repl.); VT. STAT. ANN. tit. 24, § 901a(b) (2005); VA.
CODE ANN. § 8.01-195.3 (2000); W. VA. CODE ANN. § 29-12A-5(b) (LexisNexis 2004 repl.); see
also State v. Chase Sec., Inc., 424 S.E.2d 591, 599-600 (W. Va. 1992) (holding that state officials,
even those outside the scope of statutory immunity, are personally immune "for official acts if
the involved conduct did not violate clearly established laws"). But see Kyllo v. Panzer, 535
N.W.2d 896, 903 (S.D. 1995) (holding South Dakota Codified Laws Section 21-32-17 unconstitu-
tional as applied "to state employees performing ministerial functions").
without fear of liability. It should therefore be no surprise that the

main line of attack on governmental immunity has been that it im-
properly undermines the obligation of the government to conform its
conduct to the rule of law. 53 Indeed, Justice Stevens, perhaps the
foremost judicial advocate of this view, has argued that "all Govern- 54
ments ... should generally be accountable for their illegal conduct."
Although sovereign immunity is sometimes defended as com-
pelled by the original understanding of governmental liability at the
time of the Constitution's framing or as a doctrine with limited prac-
tical significance, 5 there have been few attempts to defend it in terms
of first principles. One of the few such efforts was undertaken by
Roderick Hills, who has argued that immunity protects the public fisc
from the improvidence of public officials who, he believes, lack suffi-
cient incentive to avoid improvident payouts of excessive damages
awards or settlements. 56 But the truth of this supposition is far from
evident. After all, elected officials have political incentives to keep
taxes low and to avoid unnecessary government expenditures. Pro-
fessor Hills develops no argument to explain why these incentives are
insufficient to restrain profligate government litigation costs, nor
does he identify any empirical evidence that governments defend liti-
gation with any less vigor than private-sector defendants. I was a su-
pervisory government lawyer for many years, and I received intense
pressure from elected officials when they believed litigation expenses
were escalating, whether justifiably or not. Professor Hills seems
quite unaware of such phenomena.
Harold Krent has offered a different justification for immunity,
claiming that it preserves the separation of powers by ensuring that 57
legislative or executive policy is not undermined by damages awards.
But this argument has a cart-before-the-horse quality: it does not ex-
plain why legislative or executive prerogatives properly include the
53 See, e.g., Mark R Brown, Weathering Constitutional Change, 2000 U. ILL. L. REV. 1091, 1101-
03; Erwin Chemerinsky, Against Sovereign Immunity, 53 STAN. L. REV. 1201, 1213-15 (2001);Jack-
son, supra note 10, at 599-603; Mark C. Niles, "NothingBut Mischief": The Federal Tort Claims Act
and the Scope ofDiscretionaryImmunity, 54 ADMIN. L. REV. 1275, 1279 (2002); Randall, supra note
11, at 100-04.
Okla. Tax Comm'n v. Citizen Band Potawatomi Indian Tribe of Okla., 498 U.S. 505, 514
(1991) (Stevens,J., concurring).
55See, e.g., Jesse H. Choper &John C. Yoo, Who's Afraid of the Eleventh Amendment? The Limited
Impact of the Court's Sovereign Immunity Rulings, 106 COLUM. L. REV. 213, 233 (2006); Alfred Hill,
In Defense of Our Law of Sovereign Immunity, 42 B.C. L. REV. 485, 497 (2001); John C. Jeffries, Jr.,
In Praise of the Eleventh Amendment and Section 1983, 84 VA. L. REV. 47, 59-68 (1988); Nelson, su-
pranote 13, at 1564.
, Roderick M. Hills, Jr., The Eleventh Amendment as a Curb on BureaucraticPower, 53 STAN. L.
REV. 1225, 1234-35 (2001).
57 Harold J. Krent, Reconceptualizing Sovereign Immunity, 45 VAND. L.
REV. 1529, 1534-39
(1992).
Feb.2007] A THEORY OF GOVERNMENTAL DAMA GES LIABILITY 815
power to engage in tortious conduct without paying the injured party

compensation. In any event, this argument does notjustify immunity
in cases when the plaintiff's injury is caused by a violation of the Con-
stitution. After all, it is considered the province of the judiciary to de-
fine the scope of legislative or executive power under the Constitu-
tion!"' The constitutional tort thus merits separate consideration. 59
B. ConstitutionalTorts
Liability for constitutional torts turns on whether the defendant is

the federal government, a state, or a unit of local government, and
whether suit is brought against the government itself or a public em-
ployee.
1. FederalLiability
Sovereign immunity bars an action seeking to recover damages

from the federal government itself for a constitutional tort.6° The
federal employee who actually committed the tort, however, is in a
different legal position. In Bivens v. Six Unknown Named Agents, 6' the
Supreme Court held that federal officials are liable for damages when
they violate the plaintiffs constitutional rights. 62 A damages remedy
is available absent what the Court regards as "special factors counsel-
ing hesitation in the absence of affirmative action by Congress, 63 or
the availability of a more limited statutory remedy nevertheless
deemed to be adequate.6 4 Constitutional tort damages are not based
See, e.g., City of Boerne v. Flores, 521 U.S. 507, 536 (1997), superseded by statute on other
grounds, Religious Land Use and Institutionalized Persons Act of 2000, Pub. L. No. 106-274, 114
Stat. 803 (codified at 42 U.S.C. §§ 2000cc to 2000cc-5 (2000)); United States v. Lopez, 514 U.S.
549, 566 (1995); United States v. Will, 449 U.S. 200, 217 (1980).
59 The phrase "constitutional tort" refers to actions seeking damages for a violation
of the
plaintiff's constitutional rights. See, e.g.,
Marshall S. Shapo, ConstitutionalTort: Monroe v. Pape,
and the FrontiersBeyond, 60 Nw. U. L. REV. 277, 323-24 (1965).
SeeFDICv. Meyer, 510 U.S. 471, 477 (1994).
403 U.S. 388 (1971).
62 Id. at 395-97. Bivens involved an alleged Fourth Amendment violation, but its holding has
since been applied to other constitutional provisions. See Carlson v. Green, 446 U.S. 14, 19-20
(1980) (recognizing a cause of action for a violation of Eighth Amendment rights by a prison
official); Davis v. Passman, 442 U.S. 228, 248-49 (1979) (recognizing a cause of action for a vio-
lation of the plaintiff's Fifth Amendment rights); Butz v. Economou, 438 U.S. 478, 504-06
(1978) (recognizing a cause of action for violations of due process and First Amendment
rights).
63 Bivens, 403 U.S. at 396; accord, e.g., United States v. Stanley, 483
U.S. 669, 682-83 (1987);
Chappell v. Wallace, 462 U.S. 296, 298 (1983).
See, e.g., Schweiker v. Chilicky, 487 U.S. 412, 424-29 (1988) (rejecting a Bivens claim for
consequential damages caused by a wrongful denial of disability benefits in light of a statutory
remedy limited to an award of benefits); Bush v. Lucas, 462 U.S. 367, 386-90 (1983) (rejecting a
Bivens claim for consequential damages caused by a wrongful demotion of federal employee in
on the presumed or intrinsic value of constitutional rights; they can

be awarded only for what
65 the common law of torts considers an actual
and compensable loss.
Individual federal officials sued for constitutional torts are enti-
fled to assert the defense of qualified immunity.6 6 Under this doc-
trine, officials enjoy immunity unless their conduct contravened
clearly established constitutional law. 7 Qualified immunity is un-
available when prior decisional law makes the illegality of an official's
conduct manifest or when the official undertakes conduct that no
reasonable person could think was constitutional despite the absence
of controlling precedent. 8 The Court has considered the risk that
the threat of personal liability will inhibit the vigorous performance
of public employees' duties sufficient justification for the defense of
qualified immunity. 69
2. State and Local Liability

States enjoy immunity from damages liability for constitutional
torts by virtue of the Eleventh Amendment, which, despite its textual
limitation to cases involving citizens of different states, is thought to
incorporate a general principle of sovereign immunity.1 The Court,
light of statutory remedies limited to backpay and retroactive seniority). Bivens liability, how-
ever, is limited to federal employees. Even when sovereign immunity has been waived, no con-
stitutional tort claim is recognized against the federal government or federal agencies because
the remedy against individual employees is thought sufficient to protect the victims of constitu-
tional torts. See Meyer, 510 U.S. at 484-86; see also Correctional Servs. Corp. v. Malesko, 534 U.S.
61, 70-72 (2001) (reasoning that, because the purpose of Bivens is to deter officers, it would be
illogical to allow a constitutional tort claim against a federal agency).
65 See Memphis Cmty. Sch. Dist. v. Stachura, 477 U.S. 299, 307 (1986); Carey v. Piphus, 435
U.S. 247, 253-58 (1978).

See Harlow v. Fitzgerald, 457 U.S. 800, 813-15 (1982). There are some officials, however,
who are entitled to absolute immunity. The President enjoys absolute immunity from damages
liability arising from the performance of his duties, members of Congress and their staffs are
absolutely immune from civil liability for legislative acts, judges and other adjudicative officials
enjoy absolute immunity for their adjudicative functions, and prosecutors enjoy absolute im-
munity for prosecutorial acts. See id. at 807.
", See, e.g., Saucier v. Katz, 533 U.S. 194, 201-02 (2001); Wilson v. Layne, 526 U.S. 603, 609
(1999); Anderson v. Creighton, 483 U.S. 635, 638-40 (1987); Davis v. Scherer, 468 U.S. 183,
190-91 (1984); Harlow, 457 U.S. at 817-19.
See Hope v. Pelzer, 536 U.S. 730, 740-44 (2002); Wilson, 526 U.S. at 615-18; Mitchell v.
Forsythe, 472 U.S. 511, 530-35 (1985).
69 See, e.g., Richardson v. McKnight, 521 U.S. 399, 407-08 (1997); Wyatt v. Cole, 504 U.S.
158, 167-68 (1992); Forrester v. White, 484 U.S. 219, 223-24 (1988); Mitchell, 472 U.S. at 525-
26.
7, "The Judicial power of the United States shall not be construed to extend to any suit in
law or equity, commenced or prosecuted against one of the United States by Citizens of another
State, or by Citizens or Subjects of any Foreign State." U.S. CONST. amend. XI.
" See, e.g., Alden v. Maine, 527 U.S. 706, 712-30 (1999); Seminole Tribe of Fla. v. Florida,
517 U.S. 44, 54 (1996). The Eleventh Amendment is also understood to preclude Congress
however, construes the Eleventh Amendment to permit Congress to

authorize damages actions against states under its power to enforce
the Fourteenth Amendment. 2 Moreover, the Eleventh Amendment
does not protect state officials from personal liability for constitu-
tional torts.73 State officials are also considered "persons" amenable
to suit for depriving the plaintiff of rights under the Constitution or
federal law under Section 1983 of Title 42 of the United States
Code. 4 But state officials sued for constitutional torts may assert
qualified immunity as a defense to their personal liability. 75
Local governments are not protected by the Eleventh Amend-
ment 76 and are amenable to suit under Section 1983. 77 Section 1983,
however, has been construed to reject vicarious municipal liability for
the constitutional torts of municipal employees acting within the
scope of their employment. 78 Instead, local governments can be held
79 own policies or customs or for the acts of munici-
liable only• for their
pal policymakers. Curiously, this is precisely the basis on which dis-
cretionary immunity is available for common-law tort liability; in that
context, the touchstone for immunity is whether the challenged deci-
sion involves a judgment rooted in government policy.80 But while
discretionary immunity applies no matter how likely government pol-
icy is to inflict compensable injury within the ambit of the common
law of torts, courts impose municipal liability for constitutional torts
when a municipal policy or custom is unconstitutional or reflects de-
from compelling states to defend damages actions for an alleged violation of federal law even in
a state court. See Alden, 527 U.S. at 745-46.
72 See, e.g., Tennessee v. Lane, 541 U.S. 509, 518-20 (2004); Nev. Dep't of Human Res. v.
Hibbs, 538 U.S. 721, 726 (2003).
73 See Hafer v. Melo, 502 U.S. 21, 29-31 (1991); Scheuer v. Rhodes, 416 U.S. 232, 237-38
(1974). Additionally, the Supremacy Clause of the Constitution prevents states from enacting
immunity laws that can be used as defenses in constitutional tort litigation. See Howlett ex rel.
Howlett v. Rose, 496 U.S. 356, 375-78 (1990); Felder v. Casey, 487 U.S. 131, 141-45 (1987);
Martinez v. California, 444 U.S. 277, 284 & n.8 (1980).
74 See Hafer, 502 U.S. at 27-29. In contrast, the Court has held that a state itself is not con-
sidered a "person" amenable to suit under Section 1983. See Will v. Mich. Dep't of State Police,
491 U.S. 58, 65-71 (1989). Section 1983 provides that
[e]very person who, under color of any statute, ordinance, regulation, custom, or usage,
of any State ... subjects, or causes to be subjected, any citizen of the United States or
other person within the jurisdiction thereof to the deprivation of any rights, privileges,
or immunities secured by the Constitution and laws, shall be liable to the party in-
jured ....
42 U.S.C. § 1983 (2000).
" See Scheuer, 416 U.S. at 239-49.
,6 See, e.g., Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 368-69 (2001); Mt.
Healthy City Sch. Dist. Bd. of Educ. v. Doyle, 429 U.S. 274, 280 (1977).
77 See Monell v. Dep't of Soc. Servs., 436 U.S. 658, 690-91 (1978).
7' See id. at 693-95.
79 See, e.g., Bd. of County Comm'rs v. Brown, 520 U.S. 397, 403-04 (1997); City of Canton v.
Harris, 489 U.S. 378, 388-90 (1989).
'o See, e.g., United States v. Gaubert, 499 U.S. 315, 323-26 (1991).
liberate indifference to constitutional violations by municipal em-

ployees.81 Local governments, moreover, may not assert the defense
of qualified immunity, 2 although they enjoy immunity from punitive
damages.8 3 Precisely the opposite liability rules apply to local officials
sued for constitutional torts: they can assert the defense of qualified
immunity,4 but are subject to awards of punitive damages for inten-
tional or reckless misconduct. 5
3. Asymmetry with Private Tort Liability

The immunity doctrines for constitutional torts have for the most
part been devised by the Supreme Court.8 6 The Court, however, has
made little effort to explain why it grants government tortfeasors de-
fenses unavailable in private tort law-and the rather slender efforts
the Court has made along these lines are strikingly unpersuasive. In
Richardson v. McKnight,7 for example, as it held that the employees of
a private firm hired to run a state prison were not entitled to assert
the defense of qualified immunity, the Court endeavored to justify
the asymmetry between private-sector and governmental immunity:
First, the most important special government immunity-producing con-
cern-unwarranted timidity [caused by fear of personal liability]-is less
likely present, or at least is not special, when a private company subject to
competitive market pressures operates a prison. Competitive pressures
mean not only that a firm whose guards are too aggressive will face dam-
ages that raise costs, thereby threatening its replacement, but also that a
firm whose guards are too timid will face threats of replacement by other
firms with records that demonstrate their ability to do both a safer and a
more effective job.
This is not to say that government employees, in their efforts to act

within constitutional limits, will always, or often, sacrifice the otherwise
effective performance of their duties. Rather, it is to say that government
employees typically act within a different system. They work within a sys-
tem that is responsible through elected officials to voters who, when they
vote, rarely consider the performance of individual subdepartments or
" See Brown, 520 U.S. at 407-11; Collins v. City of Harker Heights, 503 U.S. 115, 123-24
(1992); Harris,489 U.S. at 386-89.
See Owen v. City of Independence, 445 U.S. 622, 638 (1980).
See City of Newport v. Fact Concerts, Inc., 453 U.S. 247, 258-70 (1981).
See Owen, 445 U.S. at 654-56.
See Smith v. Wade, 461 U.S. 30, 56 (1983).
The Court has sometimes derived the immunity rules for constitutional torts from the
governmental immunity and liability principles in effect when Section 1983 was enacted, see,
e.g., Buckley v. Fitzsimmons, 509 U.S 259, 268 (1993), but it has altered these rules if they come
to be viewed as counterproductive or otherwise inadvisable, see, e.g., Harlow v. Fitzgerald, 457
U.S. 800, 815-19 (1982).
87 521 U.S. 399 (1997).
civil servants specifically and in detail. And that system is often character-
ized by multidepartment civil service rules that, while providing em-
ployee security, may limit the incentives or the ability of individual de-
partments or supervisors flexibly to reward, or to punish, individual
employees. Hence a judicial determination that "effectiveness" concerns
warrant special immunity-type protection in respect to this latter (gov-
ernmental) system does not prove its need in respect to the former.
Consequently, we can find no special immunity-related need to encourage
vigorous performance.
Second, privatization helps to meet the immunity-related need to en-
sure that talented candidates are not deterred by the threat of damages
suits from entering public service.... [Ilnsurance increases the likeli-
hood of employee indemnification and to that extent reduces the em-
ployment-discouraging fear of unwarranted liability potential applicants
face. Because privatization law also frees the private prison-management
firm from many civil service law restraints, it permits the private firm,
unlike a government department, to offset any increased employee liabil-
ity risk with higher pay or extra benefits. In respect to this second gov-
ernment-immunity-related purpose then, it is difficult to find a special
need for immunity, for the guards' employer can operate like other pri-
vate firms; it need not operate like a typical government department.
Third, lawsuits may well "distrac[t]" these employees "from
their.., duties," but the risk of "distraction" alone cannot be a sufficient
grounds for an immunity.""
The Court accordingly justified special rules of governmental im-
munity on the ground that a private firm's employees are not likely to
be overdeterred by the threat of liability because they will be indem-
nified, but it offered no support for its claim that indemnification is
less likely to exist in the public sector. In fact, indemnification is
nearly always offered to public employees by statute, policy, or collec-
tive bargaining agreement,89 a fact well known to the author of
Richardson, since he recited it in another case less than two months
before Richardson came down. 90
The ubiquity of public employee indemnification should be un-
surprising; labor economics teaches that employers must offer suffi-
'8 Id. at 409-11 (internal citations and quotation marks omitted).

See, e.g., Richard Emery & Ilann Margalit Maazel, Why Civil Rights Lawsuits Do Not Deter Po-
lice Misconduct: The Conundrum and a Proposed Solution, 28 FORDHAM URB. L.J. 587, 587, 590-96
(2000); Neal Miller, Less-than-Lethal Force Weapowy: Law Enforcement and CorrectionalAgency Civil
Law Liabilityfor the Use of Excessive Force, 28 CREIGHTON L. REv. 733, 749-52 (1995); Cornelia T.L.
Pillard, Taking Fiction Seriously: The Strange Results of Public Officials' Individual Liability under
Bivens, 88 GEO. L.J. 65, 76-77 (1999); Martin A. Schwartz, Should Juries Be Informed that Munici-
pality Will Indemnify Officer's § 1983 Liability for Constitutional Wrongdoing?, 86 IOWA L. REv. 1209,
1216-23 (2001); Nicole G. Tell, Note, Representing Police Officers and Municipalities: A Conflict of
Interestfor a Municipal Attorney in a § 1983 Police Misconduct Suit, 65 FoRDHAM L. REV. 2825, 2836
(1997). For a list of state indemnification statutes, see supra note 51.
o See Bd. of County Comm'rs v. Brown, 520 U.S. 397, 436 (1997) (Breyer,J., dissenting).
JOURNAL OFCONSTITUTIONAL LAW [Vol. 9:3
cient compensation to account for the risk of liability that employees

face and that they will choose indemnification as the simplest way to
minimize the risk that their employees' financial interests in avoiding
liability will reduce their productivity. 91 There is also little reason to
credit the Court's assertion in Richardson that public employers can-
not cope with overdeterrence-public managers are accountable to
elected officials, who have political incentives to enhance the per-
formance of governmental agencies. Indeed, the Court did not iden-
tify any civil-service rules which prevent public employers from firing
those who engage in tortious misconduct or rewarding those who
improve agency performance. 92
Thus we are left with little in the way of a rationale for special
rules of governmental immunity. 93 What is more, if we credit the
Court's view that the voters have little ability to evaluate the perform-
ance of governmental agencies and that public-sector supervisors
have little control over their subordinates, then the rationale for gov-
ernmental damages liability itself is thrown into doubt. If neither the
voters nor supervisors have much ability to punish public employees
who commit torts, and given the ubiquity of indemnification of
wrongdoers and that government can readily recoup its legal ex-
penses by levying taxes, it is hard to see any efficiency or corrective-
justice justification for governmental tort liability-it neither deters
wrongdoing nor shifts costs to culpable parties. At most, governmen-
tal liability offers an injured party compensation, but if that is its only
virtue, surely this objective could be achieved through a system of
publicly funded insurance with far lower transaction costs than those
inhering in a system of tort liability. 94 Indeed, if loss-spreading is the
only objective achieved by governmental damages liability, it is far
from clear why insurance must be publicly funded at all; those con-
9, See, e.g., Alan 0. Sykes, The Economics of Vicarious Liability, 93 YALE L.J. 1231, 1239-43
(1984).
" Justice Scalia made many of these points in his dissenting opinion. See Richardson, 521
U.S. at 418-21 (Scalia, J., dissenting). For additional criticism of Richardson, see Clayton P. Gil-
lette & Paul B. Stephan, Richardson v. McKnight and the Scope oflmmunity After Privatization, 8
SUP. CT. ECON. REV. 103 (2000).
9' An additional justification sometimes offered for qualified immunity is that it reduces the
costs of adopting a new rule of constitutional law by immunizing the government from damages
liability for conduct preceding adoption of the new rule. See John C. Jeffries, Jr., The Right-
Remedy Gap in Constitutional Law, 109 YALE LJ. 87 (1999). But the more straightforward way to
address this concern is through the law of retroactivity, and on that score the Supreme Court
has held that new rules of constitutional law are applicable to all cases pending at the time the
new rule is announced. See Harper v. Va. Dep't of Taxation, 509 U.S. 86, 96-97 (1993). More-
over, municipalities are denied qualified immunity even when their policies do not violate
clearly established law. See supratext accompanying note 82.
'4 This point, for example, is at the core of the case for no-fault automobile insurance. See,
e.g., RicHARD A. POSNER, ECONOMIC ANALYSIS OF LAW § 6.14 (4th ed. 1992).
cerned about the risks of uncompensated loss could simply purchase

insurance in the private market.
C. Takings
The Fifth Amendment's Takings Clause declares that "private
property [shall not] be taken for public use, without just compensa-
tion. '9 5 Since the Fifth Amendment's text mandates the provision of
just compensation for anything that amounts to a "taking," presuma-
bly no rules of immunity from the obligation to provide compensa-
tion can be reconciled with the Fifth Amendment itself. This is
largely an academic question, however, since the federal government
has waived whatever sovereign immunity it might otherwise have
against damages liability for an uncompensated taking, 6 and it is set-
tled that a plaintiff may obtain injunctive relief against state officials
for an uncompensated taking. 97 Thus, the ability of a plaintiff to ob-
tain a remedy for an allegedly unconstitutional taking of property is
not in doubt.s
The text of the Takings Clause resolves the questions that vex
governmental liability for common-law and constitutional torts.
Unlike any other constitutional text, the Takings Clause by its terms
imposes an obligation to pay compensation. Accordingly, courts
properly award compensation for takings without worrying about the
risk of overdeterrence that concerned the Court in Richardson.
Moreover, the Takings Clause suggests a rationale for a taxpayer-
funded just-compensation requirement quite apart from any concep-
tion of deterrence. Instead of treating takings for a public use as a
9 U.S. CONST. amend. V, cl.3.

9 See, e.g., United States v. Causby, 328 U.S. 256, 267 (1946).
17 See Tindal v. Wesley, 167 U.S. 204, 212-23 (1897). This rule is an aspect of the familiar
doctrine, most frequently associated with the decision in Ex parte Young, 209 U.S. 123 (1908),
that the Eleventh Amendment does not bar an action seeking injunctive relief against a state
official to halt a violation of federal law. See, e.g.,
Frew ex rel. Frew v. Hawkins, 540 U.S. 431, 437
(2004); Verizon Md. Inc. v. Pub. Serv. Comm'n, 535 U.S. 635, 645-48 (2002); Hafer v. Melo,
502 U.S. 21, 30-31 (1991); see also United States v. Lee, 106 U.S. 196, 222-23 (1882) (stating a
similar rule authorizing injunctive relief against federal officials for an uncompensated taking
prior to the federal waiver of sovereign immunity).
' The one exception is a plaintiff seeking a remedy for a temporary uncompensated taking
against a state. Injunctive relief would be moot in such a case, but the Constitution still requires
compensation. See First English Evangelical Lutheran Church of Glendale v. County of L.A.,
482 U.S. 304, 318-22 (1987). It is unclear whether the Eleventh Amendment, as presently con-
strued, would protect a state against a suit seeking compensation under the Fifth Amendment
for either a temporary or permanent taking. For discussion of this issue, see Eric Berger, The
Collision of the Takings and Sovereign Immunity Doctrines, 63 WASH. & LEE L. REv. 493 (2006) (argu-
ing that the Takings Clause trumps state sovereign immunity), and Richard H. Seamon, The
Asymmetry of State Sovereign Immunity, 76 WASH. L. REv. 1067 (2001) (contending that the State
and state officials enjoy immunity from just-compensation suits in federal court, but not in their
own courts of general jurisdiction by virtue of the Due Process Clause).
JOURNAL OF CONSTT'UTIONAL LAW [Vol. 9:3
legal wrong in the sense that torts are legal wrongs that demand de-
terrence or corrective justice, the Takings Clause acknowledges the
propriety of taking private property for public use, while adding that
it is 'just" to expect the public to provide compensation for the
forced acquisition of property for its own use. Thus, the compensa-
tion requirement responds to a concept of unjust enrichment and
undue burden by requiring that the public compensate those whose
property it takes for public use.9 Common-law and constitutional
torts, in contrast, are accompanied by no similarly textually based
constitutional obligation to provide compensation-indeed, the Ap-
propriations Clause and its state counterparts decisively defeat any
supposed obligation along those lines-and we have seen that there
is great doubt that governmental tort liability offers either deterrence
or punishment of culpable parties, at least on the view taken in
Richardson. Thus, the contrast between governmental liability for tak-
ings and governmental liability for common-law and constitutional
torts only reinforces the problematic character of the latter. Since
the Constitution itself does not require compensation except for tak-
ings, surely we should expect some persuasive justification for a con-
stitutional or common-law tort damages remedy against government.
Such ajustification, however, has proven elusive.
II. THE POLITICS OF GOVERNMENTAL TORT LIABILITY

The emerging consensus among legal scholars is that the justifica-
tion for governmental tort liability is essentially incoherent. I hope to
demonstrate that the emerging consensus is quite wrong. But first, a
look at this emerging consensus is in order.
A. The Case Against Governmental Tort Liability

The case against governmental tort liability is straightforward-
the premises and policies of tort law fit private tortfeasors far better
than government.
1. Theories of Tort Liability

The two major schools of thought about tort law share the objec-
tive of shifting losses to culpable parties; hence both distinguish tort
Professor Michelman has provided the leading account of the just-compensation re-
quirement as rooted in a conception ofjustice framed in utilitarian terms, by using a compensa-
tion requirement to deter inefficient takings, and in Rawlsian terms, as an obligation of fairness
to individuals put to undue burdens for the sake of others. Frank I. Michelman, Property, Utility,
and Fairness: Comments on the Ethical Foundations of "Just Compensation" Law, 80 HARV. L. REV.
1165 (1967).
Feb. 2007] A THEORY OF GOVERNMENTAL DAMA GES LIABILITY 823
law from mere loss-spreading through insurance by means of a con-

ception of culpability. The instrumental account justifies tort liability
as creating an incentive to make cost-justified investments in safety."°
Thus, as Learned Hand famously put it, tort liability turns on whether
the cost of the injury multiplied by the likelihood that it would occur
exceeds the cost that the defendant would have had to incur to avoid
the loss.' 0' An employer's vicarious liability for the torts of its employ-
ees committed within the scope of their employment is therefore ne-
cessitated by the risk that employees will prove judgment proof, lead- 10 2
ing to suboptimal investments in safety absent employer liability.
The advocates of corrective justice, in contrast, argue that tort law
embodies a widely accepted moral obligation on the part of a wrong-
doer to make the injured party whole. 1 Vicarious employer liability
is warranted because an employee acting within the scope of em-
ployment is part of a larger organization that can fairly be held 01 4 re-
sponsible for the conduct of an employee working on its behalf.
The justifications for constitutional tort liability are no different;
the Supreme Court tells us that the law of constitutional torts is gov-
erned by the same principles as ordinary tort law.'10 Constitutional
tort damages have an instrumental justification in that they are
thought to create an economic incentive for the government and its
officials to make cost-justified investments in preventing constitu-
tional violations'0 6 and have a corrective-justice justification based on
an asserted moral entitlement to compensation when one has been
the victim of a constitutional wrong.'07
Essentially the same justifications are advanced for the obligation
to compensate for takings. The instrumental account contends that
the Takings Clause promotes social welfare by giving government a
disincentive to take property when the social benefits of the taking
" See, e.g., GUIDO CALABRESI, THE COSTS OF ACCIDENTS: A LEGAL AND ECONOMIC ANALYSIS
135-73 (1970); POSNER, supra note 94, at § 6.1; STEVEN SHAVELL, ECONOMIC ANALYSIS OF
ACCIDENT LAW 5-32 (1987).
0o See United States v. Carroll Towing Co., 159 F.2d 169, 172 (2d Cir. 1947).
012See, e.g., POSNER, supra note 94, at § 6.8; Alan 0. Sykes, The Boundaries of Vicarious Liability:
An Economic Analysis of the Scope of Emplyment Rule and Related Legal Doctrines, 101 HARV. L. REV.
563, 608-09 (1988).
"3 See, e.g., JULES L. COLEMAN, RISKS AND WRONGS 303-85 (1992); ERNESTJ. WEINRIB, THE
IDEA OF PRIVATE LAW 145-70 (1995); George P. Fletcher, Fairness and Utility in Tort Theory, 85
HARV. L. REV. 537, 571-73 (1972).
"" See, e.g., WEINRIB, supra note 103, at 185-87.
05 See, e.g., City of Monterey v. Del Monte Dunes at Monterey, Ltd., 526 U.S. 687, 709-10
(1999); Heck v. Humphrey, 512 U.S. 477, 483 (1994).

106 See, e.g., PETER H. SCHUCK, SUING GOVERNMENT: CITIZEN REMEDIES FOR OFFICIAL WRONGS
16-19, 135-46 (1983).

017See, e.g., Sheldon Nahmod, ConstitutionalDamages and CorrectiveJustice: A Different View, 76
VA. L. REV. 997, 1009-22 (1990).

will not exceed the cost of paying compensation to the owner. 0 8 The
intrinsic account of the just-compensation requirement contends that
it embodies a moral entitlement of property owners not to be sub-
jected to disproportionate burdens for the benefit of others.'0 9
2. The Inapplicability of the Conventional Theory to Government

Initially, the assumption that governmental tort liability works in
the same manner as the common-law liability of private tortfeasors
went unquestioned. The Supreme Court repeatedly opined that con-
stitutional tort liability can be counted upon to provide compensa-
tion while deterring official misconduct in the same manner as pri-
vate tort law," 0 and the leading commentators indulged the same
assumption."'
The first to attack this citadel was Daryl Levinson, who observed
that, unlike a private firm, the government is organized not to maxi-
mize profits but to respond to the preferences of voters, who "are not
uniquely interested in maximizing the profits, or total wealth, of the
jurisdiction."112 Instead, "government responds to political, not mar-
ket, incentives [and] cares not about dollars, only about votes."' 13 Ac-
cordingly, requiring government to pay damages for constitutional
torts will fail to reliably deter constitutional violations, which fre-
quently produce political benefits."14 For example, a policy of ran-
domly searching young men in high-crime areas could greatly in-
crease constitutional-tort liability, but it could also pay such
handsome political dividends that liability would have no deterrent
effect on elected officials." 5 Moreover, a majoritarian theory of po-
litical behavior predicts that, " [ s] o long as the social benefits of con-
stitutional violations exceed the compensable costs to the victim and
'08 See, e.g., POSNER, supra note 94, at 58-59; Michael A. Heller & James E. Krier, Deterrenceand
Distributionin the Law of Takings, 112 HARV. L. REV. 997, 997 (1999); Michelman, supra note 99,
at 1214-18.
l See, e.g., Michelman, supra note 99, at 1218-24.
"0 See, e.g., Elder v. Holloway, 510 U.S. 510, 514-15 (1994); Memphis Cmty. Sch. Dist. v.
Stachura, 477 U.S. 299, 307 (1986); Carlson v. Green, 446 U.S. 14, 20-21 (1980); Owen v. City
of Independence, 445 U.S. 622, 651-52 (1980); Butz v. Economou, 438 U.S. 478, 506 (1978);
Robertson v. Wegmann, 436 U.S. 584, 590-91 (1978).
"' See, e.g., John C. Jeffries, Jr., Damagesfor Constitutional Violations: The Relation of Risk to In-
jury in ConstitutionalTorts, 75 VA. L. REV. 1461, 1461-64 (1989); Larry Kramer & Alan 0. Sykes,
Municipal Liability Under § 1983: A Legal and Economic Analysis, 1987 Sup. CT. REV. 249, 276-87;
Daniel J. Meltzer, DeterringConstitutional Violations by Law Enforcement Officials: Plaintiffs and De-
fendants as PrivateAttorneys General,88 COLUM. L. REV. 247, 253-78 (1988).
11 Daryl J. Levinson, Making Government Pay: Markets, Politics,
and the Allocation of Constitu-
tional Costs, 67 U. CHI. L. REv. 345, 355 (2000).
Id. at 420.
Id. at 367-68.
15 Id. at 369-70.
are enjoyed by a majority of the population, compensation will never

deter a majoritarian government from violating constitutional
rights.. ,,."- As for the obligation to compensate for takings, Levin-
son argued that compensation is unlikely to create an incentive to
engage in only efficient takings: "why should we expect government
to fully internalize the benefits of takings when it does not receive
them in the form of revenues?" 7 In particular, even an inefficient
taking could garner majority support when a majority of voters stands
to experience benefits from it that exceed the increase in taxes to
fund compensation. 8 Levinson added that public-choice theory
predicts that powerful interest groups might be able to successfully
lobby for unconstitutional policies or improvident takings that pro-
vide them with special benefits despite the obligation to pay compen-
sation; an influential homeowners association, for example, might
lobby to condemn nearby property it dislikes even though the cost of
condemnation might exceed its overall benefits." 9 Levinson con-
cluded, "the most consistent prediction generated by interest group
analysis is that compensation for takings or constitutional torts will
tend to defuse political opposition and therefore increase the inci-
dence of both."' At best, "any predictions about the incentive effects
of constitutional cost remedies on government behavior are highly
suspect."21
Levinson also rejected corrective justice as a justification for gov-
ernmental liability for constitutional torts or a requirement of com-
pensation for takings. As to the former, the fact that "constitutional
tort compensation ultimately comes from the pockets of taxpay-
ers... attenuates the connection between moral responsibility and
the burden of rectification.', 2 One could add that the shareholders
6 Id. at 370. Levinson added that deterrence cannot be expected when a majority of citi-
..
zens experience greater benefits than any increase in their taxes. Id. at 370-73.
17 Id. at 350 (footnote omitted).
.. Id. at 364-66. He illustrated the point:
[T]he proposed regulation will benefit each of citizens 1-6 (the members of the mini-
mum winning coalition) by $2000, while costing each of citizens 7-10 $4000. Citizens 1-
6 will support the regulation because it leaves them better off by $400 ($2000 in direct
benefits minus one-tenth of $16,000, or $1600, in compensation taxes), and the ineffi-
cient regulation will therefore pass.
Id. at 365.
19 Id. at 375-79.
Id. at 379-80.
2 Id. at 386-87; see also Daryl J. Levinson, Empire-Building in Constitutional Law, 118 HARV. L.
REV. 915, 964-68 (2005) (contending that government actors care significantly more about po-
litical incentives than about budgetary considerations).
i Levinson, supra note 112, at 408. Levinson also observed that many constitutional viola-
tions are systemic in nature, making it unrealistic to identify the particular plaintiff as a victim,
id. at 409, and that many constitutional injuries are difficult to monetize in any fair or princi-
pled fashion, id. at 410.
of private corporations have the benefit of liability limited to their in-

vestment even though they can readily sell their stock at any time, yet
taxpayers must fund essentially unlimited liability and face substantial
costs if they wish to "exit" the jurisdictions that tax them to fund gov-
ernmental liabilities.1 2 3 As for takings, Levinson rejected a moral
claim of property owners to compensation on the ground that the 24 ex-
isting distribution of property cannot itself be defended as just.
Although he seemed not to realize it, Levinson's view argues
against governmental liability for common-law torts no less than for
constitutional torts and takings. On Levinson's account, there is no
instrumental justification for governmental liability for common-law
torts since we cannot know that government will make cost-justified
investments in loss prevention to minimize its future liability. When
the political cost of diverting public resources to loss prevention is
sufficiently high, government will not make the investment even
when it is economically justified. As a result, tort liability cannot be
expected to promote efficient governmental investment in loss pre-
vention. And since the economic cost of damages awards falls on
taxpayers not responsible in any direct fashion for tortious conduct,
the corrective-justice rationale for governmental damages liability for
common-law torts is also wanting. At most, we are left with an argu-
ment for providing those injured by tortious government conduct
with some form of publicly funded insurance-although, as we have
seen, the justification for having taxpayers fund this obligation rather
than leaving the insurance decision to each individual is entirely un-
clear.
Levinson is clearly onto something. While we can assume that
private-sector tortfeasors are primarily motivated to maximize reve-
nues and accordingly are likely to experience a form of deterrence or
punishment from tort liability, voters' demands on government are
much broader and more amorphous. Accordingly, the deterrent and
punitive effects of tort liability are far more uncertain in the public
sector. 1 Indeed, as Levinson points out, tort liability may even pay
For a discussion of the nature of limited shareholder liability, see, for example, Janet Coo-
per Alexander, Unlimited Shareholder Liability Through a ProceduralLens, 106 HARV. L. REV. 387
(1992).
1' See Levinson, supra note 112, at 397-400.
'2' To be sure, there are those that doubt the efficacy of tort liability even in the private sec-
tor in terms of deterrence or corrective justice. See generally Richard L. Abel, A Critique of Torts,
37 UCLA L. REv. 785 (1990) (suggesting that tort law is inefficient with regard to its purported
goals of responding to the victim's need for compensation and encouraging future safety);John
A. Siliciano, Corporate Behavior and the Social Efficiency of Tort Law, 85 MICH. L. REv. 1820 (1987)
(discussing how various doctrines of corporate law undermine the practical utility of the social
efficiencyjustification of torts); Stephen D. Sugarman, DoingAway with Tort Law, 73 CAL. L. REV.
555 (1985) (analyzing the ineffectiveness of tort as a deterrent and critiquing the compensation
system of tort). Levinson's argument, however, does not rest upon an assessment of the efficacy
Feb. 2007] A THEORY OFGOVERNMENTAL DAMAGES LIABILITY 827
political dividends in the public sector. Thus, after Levinson's attack,

the justification for any form of governmental damages liability law is
in shambles.
The efforts made to date to defend governmental tort liability
from Levinson's attack have been deeply unsatisfying. For example,
Bernard Dauenhauer and Michael Wells have defended constitu-
tional tort liability in terms of corrective justice, arguing that the tax-
payers are properly held responsible for governmental torts since
they largely correspond to the universe of voters responsible for elect-
ing those who run the government. 21 6 But this is hardly corrective jus-
tice from the standpoint of those voter-taxpayers who backed the los-
ing candidate in the last election. Moreover, a shareholder who is
dissatisfied with management policy may always sell, but a taxpayer
can relocate to another jurisdiction only at considerable cost. And
taxpayers do not have the kind of effective control over government
policy vested in owner-shareholders. Voting is governed by a one-
person-one-vote rule,127 but taxes are not levied on that basis. 28 To
top it off, the responsibility of politically accountable elected officials
for most governmental torts is unclear. Courts will usually impose
tort liability without any requirement that elected officials were in-
volved in or culpable for the tort; only a Section 1983 claim against a
municipality requires any showing of culpability on the part of poli-
cymakers. 129 As the Supreme Court explained in Richardson, however,
most constitutional torts are committed by relatively low-level officials
subject to limited control by elected officials or the voters. 30 Thus,
even if taxpayers are fairly held accountable for the conduct of
elected officials, the responsibility of elected officials for most gov-
ernmental torts that give rise to damages liability is highly attenuated.
Attacks on Levinson's position from an instrumental perspective
have been no more persuasive. For example, Mark Brown has argued
that citizens are more likely to obey the law when the government is
required to honor its own legal obligations and is held liable for its
violations.13 ' He offers no empirical evidence to support this view,
however, and his theory seems implausible. The Supreme Court rec-
of either public or private tort liability; his point is that whatever the efficacy of tort liability in
the private sector, it has sigfinicantly less efficacy in the public sector.
'26 Bernard P. Dauenhauer & Michael L. Wells, CorrectiveJustice and Constitutional Torts, 35
GA. L. REV. 903, 914-28 (2001).
12 See, e.g., Bd. of Estimate v. Morris, 489 U.S. 688, 692 (1989); Avery v. Midland County, 390
U.S 474 (1968).
' Indeed, it is usually unconstitutional to make eligibility to vote turn on one's status as a
taxpayer. SeeKramerv. Union Free Sch. Dist. No. 15, 395 U.S. 621, 632 (1969).
29 See supra Part I.
,0 See supra text accompanying note 88.

Mark R. Brown, DeterringBully Government: A Sovereign Dilemma, 76 TUL. L. REV. 149, 172-
78 (2001).
JOURNAL OF CONSTTFUTIONAL LA W [Vol. 9:3
ognized constitutional tort liability as we know it today in Monroe v.

Pape.132 While constitutional tort litigation exploded in the wake of
that decision, 133 there is no evidence that people became more law
abiding. In fact, crime rates rose in the years following Monroe, and
criminologists certainly have not observed any relationship between
the rise of 4 constitutional tort litigation and norms of law-
abidingness.1
Somewhat more plausibly, Myriam Gilles has attempted to meet
Levinson on his own terms. Admitting that governmental tort liabil-
ity can have little deterrent effect unless it exacts a political price, she
claims that tort litigation does just that by unearthing damaging in-
formation or producing adverse verdicts and rulings, thus creating
damaging publicity for officeholders. 3 5 But she offers a paucity of
empirical evidence to support this claim. Especially in light of the
enormous scrutiny of government-systemically promoted by a com-
petitive political system and a free press-there is little reason to be-
lieve that civil litigation is a major vehicle for unearthing governmen-
tal misconduct. Professor Gilles offers no evidence that damages
litigation has produced effective political accountability; indeed, it is
difficult to think of any major governmental scandal uncovered by
tort litigation. The major national scandals of recent decades, such
as Watergate or the Iran-Contra affair, for example, were not un-
earthed by tort plaintiffs. 36 At the local level, it is equally difficult to
identify a major scandal unearthed by a tort plaintiff; it was a Los An-
geles police investigation, for example, that discovered the huge
12 365 U.S. 167, 183 (1961), overruled in part on other grounds by Monell v. Dep't of Soc. Servs.,
436 U.S. 658 (1978).
...As Justice Powell once observed, "[i]n 1961 ... only 270 civil rights actions were begun in
the federal district courts[, but by] 1981[] over 30,000 such suits were commenced." Patsy v.
Bd. of Regents, 457 U.S. 496, 533 (1981) (Powell,J., dissenting). To be sure, these figures are
somewhat misleading because they include cases that are not fairly characterized as constitu-
tional torts, such as employment discrimination actions. See Theodore Eisenberg & Stewart
Schwab, The Reality of Constitutional Tort Litigation, 72 CORNELL L. REv. 641, 658-68 (1987). Still,
there can be little doubt that, since Monroe, constitutional tort litigation has expanded dramati-
cally.
' See generally GARY LAFREE, LOSING LEGITIMACY: STREET CRIME AND THE DECLINE OF SOCIAL
INSTITUTIONS IN AMERICA (1998) (analyzing the increase in American crime rates from the end
of World War II to the early 1990s, particularly in the 1960s).
' Myriam E. Gilles, In Defense of Making Government Pay: The Deterrent Effect of Constitutional
Tort Remedies, 35 GA. L. REV. 845, 859-65 (2001). SimilarlyJames Park has argued that awards
of damages are necessary so that courts are able to articulate constitutional principles and rights
in cases where other types of relief are unavailable. James J. Park, The Constitutional Tort Action
as Individual Remedy, 38 HARv. C.R.-C.L. L. REV. 393, 422-24 (2003).
" See CARL BERNSTEIN & BOB WOODWARD, ALL THE PRESIDENT'S MEN (1974) (Watergate); 1
LAWRENCE E. WALSH, FINAL REPORT OF THE INDEPENDENT COUNSEL FOR IRAN/CONTRA MATTERS
23-25 (1993) (Iran-Contra).
Feb. 2007] A THEORY OFGOVERNMENTAL DAMAGES LIABILr"'Y 829
Ramparts police corruption scandal. 37 What is more, only a tiny frac-

tion of civil litigation against the government is likely to generate
publicity, much less significant political consequences. When a gov-
ernmental tort has political consequences, it is more likely because of
the underlying conduct than the ensuing litigation. For example,
Rodney King's beating by Los Angeles police officers produced an
immediate and enormous adverse public reaction well before any
civil litigation arose.' 38 Given the incentive of the political opposition
and the press, among others, to unearth governmental misconduct,
and the ordinary political checks that come into play whenever alle-
gations of governmental misconduct enter the political arena, the
marginal utility of civil litigation in exposing such misconduct and
punishing officeholders is not likely to be great.
But if Professor Gilles is right, then her argument suggests that
the cure may be worse than the disease. If civil litigation is uniquely
valuable at ferreting out governmental misconduct because of the fi-
nancial incentive that the civil plaintiff has to pursue such allegations,
then elected officials would presumably pay premiums-beyond the
amount necessary to compensate the plaintiff-in order to settle liti-
gation that might otherwise cause political embarrassment. 39
Thus,
governmental tort liability effectively enables plaintiffs to extort ex-
cessive settlements funded with tax dollars. Indeed, we have seen
that to avoid
4 1
just this perverse result Professor Hills defends sovereign
immunity.
'" Andrew Chung, 'O'er the Ramparts... Still Gallantly Streaming?', CHAMPION, Apr. 2000, at
12.
" See, e.g.,
Seth Mydans, Tape of Beating Revives Charges of Racism, N.Y. TIMES, Mar. 7, 1991, at
A18; Elaine Woo, Rev. JacksonJoins Callfor Gates' Ouster, Scolds Bradley, L.A. TIMES, Mar. 17, 1991,
at BI.
'9 Publicity surrounding settlements can be minimized through sealed settlement agree-
ments, which are used with increasing frequency by defendants who wish to inhibit public scru-
tiny of the events surrounding the litigation. See, e.g., Alan F. Blakley, To Squeal or Not To Squeal:
Ethical Obligations of Officers of the Court in Possession of Information of Public Interest, 34 CUMB. L.
REV. 65, 79-86 (2003); Laurie Kratky Dor, Secrecy By Consent: The Use and Limits of Confidentiality
in the Pursuitof Settlement, 74 NOTRE DAME L. REV. 283, 384-401 (1999); Heather Waldbeser &
Heather DeGrave, A Plaintiffs Lawyer's Dilemma: The Ethics of Entering a Confidential Settlement, 16
GEO.J. LEG. ETHICS 815, 815-20 (2003).
"o See supra text accompanying note 56. Christopher Serkin has offered a defense of the tak-
ings liability of local governments on the ground that local governments are largely funded by
property taxes and property owners largely internalize the costs and benefits of local land-use
policy. Christopher Serkin, Big Differences for Small Governments: Local Governments and the Tak-
ings Clause, 81 N.Y.U. L. REV. 1624, 1644-65 (2006); see also WILLIAM A. FISCHEL, THE
HOMEVOTER HYPOTHESIS: HOW HOME VALUES INFLUENCE GOVERNMENT TAXATION, SCHOOL
FINANCE, AND LAND-USE POLICIES 39-71 (2001) (arguing that residents internalize the costs and
benefits of local-government activities because their property tax payments represent the cost of
local government and the quality of local-government service is reflected in local property val-
ues). We have seen, however, that the universes of voters and property owners differ, and for all
but the broadest-based public projects, only a small minority of property owners will be likely to
JOURNAL OF CONSTFUTIONAL LAW [Vol. 9:3
These responses, however, are far from the typical academic reac-
tion to Professor Levinson's critique of governmental damages liabil-
ity. It is perhaps only a slight exaggeration to say that Levinson has
revolutionized academic thinking about governmental damages li-
ability. 41 Most academics have been persuaded by Levinson; it has
now become fashionable to warn that the consequences of imposing
damages liability on government are uncertain at best.14 ' For this rea-
son, some scholars now advocate imposing personal liability on pub-
lic employees for their torts while mandating sanctions against such
employees.'43 Sanctioning an individual official, however, ignores the
very dynamic that has led to a nearly universal regime of indemnifica-
tion-if individual officials face a credible threat of serious sanctions,
the risk of overdeterrence would be quite real, and the resulting de-
cline in the effective value of public-sector compensation would pro-
duce a concomitant decline in the quality of the public workforce
experience any perceptible change in the value of their property as a result of the taking. Thus,
it is far from clear that Professor Serkin has persuasively answered Levinson, even in the limited
context of local-governmental takings liability.
" Indeed, the editors of the Georgia Law Review found Levinson's argument so provocative
that they devoted a symposium to an assessment of Levinson's article. Symposium, Re-examining
First Principles: Deterrenceand CorrectiveJustice in ConstitutionalTorts, 35 GA. L. REV. 837 (2001).
112 For a survey of some of the many articles that refer to Levinson's argument
and the uncer-
tain effects of the imposition of damages liability on governments, see, for example, Vicki Been
& Joel C. Beauvais, The Global Fifth Amendment? NAFTA 's Investment Protections and the Misguided
Quest for an International "Regulatory Takings" Doctrine, 78 N.Y.U. L. REV. 30, 89-90 (2003);
Choper & Yoo, supra note 55, at 259; Reza Dibadj, Reconceiving the Firm, 26 CARDOZo L. REV.
1459, 1517 (2005); Alexandra White Dunahoe, Revisiting the Cost-Benefit Calculus of the Misbehav-
ingProsecutor: Deterrence Economics and Transitory Prosecutors, 61 N.Y.U. ANN. SURV. AM. L. 45, 58-
59 (2005); Nicole Stelle Garnett, The Public-Use Question as a Takings Problem, 71 GEO. WASH. L.
REv. 934, 956-61 (2003); Heather K. Gerken, A Third Way for the Voting Rights Act: Section 5 and
the Opt-in Approach, 106 COLUM. L. REV. 708, 721 n.43 (2006); Clayton P. Gillette, Kelo and the
Local Political Process, 34 HOFSTRA L. REV. 13, 15-16 (2005); Louis Kaplow, Transition Policy: A
Conceptual Framework, 13 J. CONTEMP. LEGAL ISSUES 161, 192-95 (2003); Eugene Kontorovich,
The Constitution in Two Dimensions: A Transaction Cost Analysis of ConstitutionalRemedies, 91 VA. L.
REv. 1135, 1187-88 (2005); Gia B. Lee, Persuasion, Transparency, and Government Speech, 56
HASTINGS L.J. 983, 1031 (2005); Evan J. Mandery, Efficiency Considerations of Compensating the
Wrongfully Convicted, 41 CRIM. L. BULL. 287-89 (2005);John T. Parry, ConstitutionalInterpretation,
Coercive Interrogation, and Civil Rights LitigationAfter Chavez v. Martinez, 39 GA. L. REV. 733, 831-
35 (2005); Richard A. Primus, Bolling Alone, 104 COLUM. L. REV, 975, 1014-15 (2004); Christo-
pher Serkin, The Meaning of Value: Assessing Just Compensationfor Regulatory Takings, 99 NW. U. L.
REV. 677, 725-27 (2005); and Henry A. Span, Public Choice Theory and the Political Utility of the
Takings Clause,40 IDAHO L. REV. 11, 22-24 (2003).
M See, e.g., Dunahoe, supra note 142, at 71-109; Emery & Maazel, supra note
89, at 596-600;
David Rudovsky, Running in Place: The Paradox of Expanding Rights and Restricted Remedies, 2005
U. ILL. L. REV. 1199, 1225-26. Another potential solution is to rely on property rather than li-
ability rules, which enforce legal rights through the issuance of injunctive relief. See Guido
Calabresi & A. Douglas Melamed, Property Rules, Liability Rules, and Inalienability: One View of the
Cathedral, 85 HARV. L. REV. 1089 (1972). Injunctive relief is unavailable, however, when the
wrongful conduct at issue has ceased. See, e.g., Renne v. Geary, 501 U.S. 312, 320-21 (1990);
City of Los Angeles v. Lyons, 461 U.S. 95, 105-06 (1983). Thus, injunctive relief will frequently
be unavailable as a remedy when the alleged misconduct is not ongoing.
Feb. 2007] A THEORY OFGOVERNMENTAL DAMAGES LIABILfTY 831
unless the government offered compensation amounting to effective

indemnification. 444 Sanctioning individual public employees is no an-
swer to Levinson. 1
When I first read Levinson's article, I was a rather senior official in
municipal government in Chicago. Levinson's claim that the gov-
ernment responds to political and not market forces struck me as ex-
actly right. I had long been editing out of briefs that came to me for
review an argument that expansive liability of police officers or fire-
fighters might cause cities to discontinue such services because of
their cost. Attorneys that I supervised had encountered this kind of
argument against tort liability in law school, where private-sector li-
ability is the focus of instruction, but try as one might to adapt it to a
municipal defendant, this claim simply was not credible in the public
sector. Private entities might get out of a business if liability became
too costly, but no politician would ever propose reducing police or
fire protection in order to avoid tort liability. At the same time, Lev-
inson's claim that governmental damages liability has indeterminate
political consequences did not ring true to me. If Levinson were
right, the officials within Chicago's government with whom I worked
should have been indifferent to municipal damages liability. Yet, in
fact, city government devoted enormous resources to trying to mini-
mize liability. Why was that so? Why did Chicago pay for risk manag-
ers and a legal department that vigorously contested liability if, as
Levinson supposed, the payment of compensation was actually an ef-
fective means of defusing political opposition?
B. The Politics of GovernmentalDamages Liability

Precisely because government responds to political and not eco-
nomic incentives, an assessment of governmental tort liability re-
quires an inquiry into the political significance of liability.
" See supra text accompanying note 91.

... In this connection, it is worth contemplating the fate of Westfall v. Erwin, 484 U.S. 292
(1988). In that case, the Court held that federal employees lack immunity from liability in tort
even when sued for acts within the scope of employment unless their conduct is discretionary.
See id. at 296-99. Within months, Congress responded by enacting the Federal Employees Li-
ability Reform and Tort Compensation Act of 1988, which amended the Federal Tort Claims
Act (FTCA) to provide that a suit against the United States under the FTCA would provide the
exclusive remedy for any plaintiff injured by the negligent or wrongful act of a federal em-
ployee. Pub. L. No. 100-694, 102 Stat. 4563 (1988); see also United States v. Smith, 499 U.S. 160,
165-69 (1991). This sequence of events illustrates the political dynamics that come into play
when government employees face personal liability.
JOURNAL OFCONSTITUTIONAL LAW [Vol. 9:3
1. A Theory of PoliticalBehavior
Consider the following account of political behavior as it relates to

governmental damages liability: The primary objective of elected of-
ficials is to prevail in their next election, whether they are seeking re-
election or some different, presumably higher office. To do this, they
offer voters and other potential supporters what they believe to be an
attractive mix of policies. In particular, elected officials impose taxes
and allocate available resources generated by those taxes in what they
regard as a politically optimal manner. They must distribute public
resources in a manner that meets the demands of actual and desired
supporters while keeping taxes low enough as to avoid politically off-
setting opposition. In this fashion, elected officials seek to craft poli-
cies that a majority coalition will conclude have created a sufficiently
high ratio of benefits to taxes paid to make success sufficiently prob-
able in the next election. On this view, it is immaterial whether one
accepts a majority-rule account of political behavior-postulating that
elected officials are concerned with the views of a majority of their
constituents-or a public-choice account asserting that elected offi-
cials are likely to be more responsive to those groups with particular
stakes in various public-policy issues. 146 Under the theory advanced
above, all that matters is that elected officials concern themselves with
utilizing the public resources over which they have control to secure a
likelihood of success in future elections.
On this view of political behavior, elected officials will be highly
sensitive to tort liability. Tort liability reduces the resources available
for allocation in order to produce what elected officials regard as the
optimal ratio of governmental benefits to taxes. To pay judgments
and other litigation expenses, officials must either raise taxes; incur
debt in jurisdictions where that is permitted, which, of course, re-
quires that additional tax revenues be diverted in future budgets to
repay the debt with market-rate interest; or experience a reduction in
the funds available for allocation in what elected officials regard as
the politically optimal manner. All of these options impose signifi-
cant political costs on incumbents. Moreover, sums paid out to
judgment creditors result in particularly small political dividends
since these plaintiffs likely view such sums as an entitlement rather
than a politically bestowed reward likely to produce loyalty to incum-
bent officials. And, because there is a political incentive to maximize
discretionary spending on politically favored constituencies while
minimizing taxes, an elected official should be willing to make in-
vestments in loss prevention in order to reduce governmental liability
costs. Indeed, the interest in maximizing political control over tax
.. See, e.g., POSNER, supra note 94, at § 19.3.

and spending policy is one all elected officials share, regardless of

partisan affiliation or whether they exercise executive or legislative
powers. Finally, this theory also predicts that there will be political
pressure-from interest groups unable to obtain desired benefits
when resources must be allocated to litigation and from elected offi-
cials themselves-to enact immunity legislation in order to respond
to the voters' tax and spending preferences.
2. Assessing the Theory

How would one go about assessing the merits of the theory ad-
vanced above? The initial premise-that politicians are concerned
about winning the next election-is perhaps uncontroversial. To be
sure, the view that elected officials behave in ways that enhance their
likelihood of success in the next election could be rejected by a "Pro-
files in Courage" model of elected officials discharging their duties
without regard for electoral consequences. There is little reason to
embrace such a model, however; the political equivalent of natural
selection will eventually oust elected officials who ignore constituent
preferences. In any event, political-science research consistently
shows that reelection is one of the primary concerns of elected offi-
cials.147
The next premise-elected officials consider their power to allo-
cate available public resources an important means of building elec-
toral support-is perhaps more doubtful. One could argue that in-
cumbent officials have other, more useful ways of building or
maintaining political support, or that they rarely fear defeat at the
next election. Incumbents, however, face one type of vulnerability
not shared by challengers: as the political-science literature demon-
strates, the voters have much more information about incumbents as
a consequence of their records, and, as a result, the efficacy of cam-
148
paign promises is reduced when they come from incumbents. For
that reason, incumbents have special incentives to utilize their con-
trol over tax and spending policy in order to offset the disadvantage
of a concrete record.
Considerable empirical evidence supports this view of the impor-
tance of allocational politics to incumbents. At the local level, as Paul
Peterson has observed, politicians have long been concerned with al-
locating resources to maximize political benefits:
,41 See, e.g., RICHARD F. FENNO, JR., CONGRESSMEN IN COMMITrEES 1-14 (1973); DAVID R.
MAYHEW, CONGRESS: THE ELECTORAL CONNECTION 13-17 (1975).
4' See, e.g.,John Ferejohn, Incumbent Performance and Electoral Control, 50 PUB. CHOICE 5, 7-8
(1986).
Many political controversies fall within the allocational arena. Loca-

tional politics, which consumes much of the energy of urban politicians,
usually involves the allocation of government resources to one or another
part of the city. Where should a school building be sited? What should
be the route of a badly needed roadway? Should a new hospital be built
on parkland? The housekeeping services of government compose an-
other set of allocational issues. What streets should be cleared first?
Which sidewalks need repairing? How often is the garbage to be col-
lected? Where should fire stations be located? What should be city pol-
icy on the enforcement of parking and traffic ordinances? Even minor
tax questions are largely allocational issues with little effect on the city's
long-term interests. Should needed revenue be collected through in-
creasing water and sewer fees, by charging admission to city museums, by
increasing library fines, or by slightly raising the property tax?
The allocational policies that have provoked the most enduring local
conflict have related to the terms and conditions of public employment.
Their centrality in local politics has often been attributed to their mate-
rial and divisible qualities. Jobs can quite literally be divided into ever
smaller packages and distributed to ever more specific segments of the
community. Policies controlling their disposition can be almost infinitely
varied from one sector of public service to another and from one rank to
another within the service. Different groups can each be given their own
parcel of positions, whose recruitment and promotion policies can be
geared to that group's interests. The benefit, moreover, is perfectly con-
crete and material,
149 a tangible good whose value can be fairly accurately
calculated.
Similarly, studies of congressional behavior consistently identify a de-
sire on the part of members of Congress to allocate public resources
to the use of "pork barrel" projects that are likely to provide particu-
larized benefits to their constituents.5 0 Studies of both executive and
legislative behavior at the state level, while less frequent, reach the
same conclusion. 5 ' Litigation costs, however, reduce the resources
"' PAUL E. PETERSON, CITY LIMITS 150-51 (1981) (internal citation omitted).
INSee, e.g., JOHN A. FEREJOHN, PORK BARREL POLITICS: RIVERS AND HARBORS LEGISLATION,
1947-1968, at 233-52 (1974); MAYHEW, supra note 147, at 52-59; ROBERT M. STEIN & KENNETH
N. BIcKERS, PERPETUATING THE PORK BARREL: POLICY SUBSYSTEMS AND AMERICAN DEMOCRACY
118-36 (1995); Steven J. Balla et al., Partisanship,Blame Avoidance, and the Distribution of Legisla-
tive Pork, 46 AM. J. POL. ScI. 515 (2002); Frances E. Lee, Senate Representation and Coalition Build-
ing in DistributivePolitics, 94 AM. POL. SC1. REV. 59 (2000); Matthew D. McCubbins & Terry Sulli-
van, Constituency Influences on Legislative Policy Choice, 18 QUALITY & QUANTITY 299 (1984);
Robert M. Stein & Kenneth N. Bickers, CongressionalElections and the Pork Barre4 56 J. POL. 377
(1994); Barry R. Weingast et al., The PoliticalEconomy of Benefits and Costs: A NeoclassicalApproach
to Distributive Politics, 89J. POL. ECON. 642 (1981); Barry R. Weingast, A Rational ChoicePerspective
on CongressionalNorms, 23 AM.J. POL. ScL. 245 (1979). For an interesting demonstration of the
efficacy of this type of allocative politics, see Steven D. Levitt & James M. Snyder, Jr., The Impact
of FederalSpending on HouseElection Outcomes, 105 J. POL. ECON. 30 (1997).
151See, e.g., ALAN ROSENTHAL, GOVERNORS AND LEGISLATURES: CONTENDING POWERS 133-35,
158-60 (1990); Charles S. Bullock, III, & M.V. Hood, III, 1When Southern Symbolism Meets the Pork
Barrel: Opportunityfor Executive Leadership,86 Soc. Sci. Q. 69, 78-83 (2005).
Feb. 2007] A THEORYOFGOVERNMENIALJDAMAGES LIABILFFY 835
available to elected officials available for allocational politics. For

that reason, elected officials have reason to be sensitive to those costs.
Even granting the concern of elected officials with maximizing the
resources available for allocational politics, objections remain to the
theory of political behavior advanced above. For example, it is rea-
sonable to believe that the time frame of concern to politicians is the
next electoral cycle and that their political judgments are therefore
made with only that time frame in mind. 5 2 For that reason, elected
officials might ignore litigation costs or liability exposure, believing
that they have no real ability to reduce them quickly enough to affect
the current electoral cycle. Still, there is reason for skepticism about
this view of the time horizons of public officials-most politicians
likely plan long careers in public service and will pay a political price
if they are still in office when tort judgments must be paid. 5
One might also argue that litigation costs are not a sufficiently
significant proportion of government budgets to be of concern to
elected officials. It is difficult to assess this objection, since it is noto-
riously difficult to obtain information from governments about litiga-
tion-related costs, although anecdotal evidence suggests that Tovern-
mental tort litigation generates substantial budgetary outlays. For
example, between 1994 and 1996, New York City paid some seventy
million dollars to plaintiffs in police misconduct litigation. 55 The
City of Los Angeles was required to expend the same amount to settle
police misconduct litigation stemming from corruption in a single
antigang unit.156 And a U.S. Department of Justice task force con-
cluded that in the early 1980s municipalities had experienced signifi-
cant financial difficulties as a result of rapidly escalating tort liability
57
costs. At the federal level, Harold Krent has estimated that discre-
tionary-function immunity saves the government several billion dol-
'52See, e.g., James M. Buchanan & Dwight R. Lee, Tax Rates and Tax Revenues in PoliticalEqui-
librium: Some Simple Analytics, 20 ECON. INQUIRY 344, 345-50 (1982).
,51Moreover, time horizons may not be all that different in the public and private sectors.
There is increasing concern that the emphasis on short-term profits and stock prices in the pri-
vate sector has incented management to be less than fully sensitive to threats to the long-term
financial health of the firm. See, e.g., Jeffrey N. Gordon, What Enron Means for the Management
and Control of the Modern Business Corporation: Some Initial Reflections, 69 U. CHI. L. REV. 1233,
1245-48 (2002); David Millon, Why Is CorporateManagement Obsessed With QuarterlyEarnings and
What Should Be Done About It?, 70 GEO. WASH. L. REv. 890, 892-900 (2002); Lawrence E.
Mitchell, A CriticalLook at Corporate Governance, 45 VAND. L. REV. 1263, 1283-94 (1992).
' See, e.g., Marc L. Miller & Ronald F. Wright, Secret Police and the Mysterious Case of the Missing
Tort Claims, 52 BUFF. L. REV. 757, 766-74 (2004).
1 Emery & Maazel, supranote 89, at 590 (citation omitted).
56 John M. Broder, Los Angeles Paying Victims $70 Million for Police Graft, N.Y. TIMES, Apr. 1,
2005, at A20.
151 See U.S. DEP'T OFJUSTICE, REPORT OF THE TORT POLICY WORKING GROUP ON THE
CAUSES,
EXTENT AND POLICY IMPLICATIONS OF THE CURRENT CRISIS IN INSURANCE AFFORDABILITY AND
AVAILABILITY 8-9 (1986).
lars per year.' 5s Even that estimate may be low. In United States v.
Gaubert,'5 for example, the Court held that discretionary immunity
barred a hundred-million-dollar claim representing the loss of a sin-
gle investor allegedly caused by federal negligence in supervising a
single federally regulated bank. 60 Had there been no immunity, the
budgetary impact of liability for negligent supervision of federally
regulated banks could have been enormous. But reliable statistics
are difficult to find; in particular, it is difficult to know how much the
government saves because of the cases that are never filed as a conse-
quence of immunity.1 62 And it is also difficult to evaluate the political
magnitude of these expenditures since their political significance de-
pends on a comparison to the portion of government budgets that
are available for discretionary distribution after fixed costs not subject
to effective political control are disregarded-a calculation that is dif-
ficult if not impossible to make.
But even if elected officials perceived litigation-related costs as po-
litically significant, one could still object that elected officials may
have little ability to understand or control litigation costs. Although
most liability-creating events are the result of the actions of relatively
low-level officials, civil-service protections and a variety of other ar-
rangements insulate the bureaucracy from political control, leaving
elected officials with limited authority over bureaucratic behavior.
This may mean that elected officials view litigation expenses as essen-
tially a fixed cost over which they have no control. Moreover, senior
bureaucrats have an incentive to conceal problems within their areas
of responsibility from elected officials.' 64 Still, elected officials must
ultimately appropriate sufficient funds to pay for litigation expenses,
so there is reason to believe 65 that elected officials at least have some
knowledge of litigation costs.
'58 Harold J. Krent, PreservingDiscretion Without SacrificingDeterrence: Federal Government Liabil-

ity in Tort, 38 UCLA L. REV. 871, 871 (1991).
499 U.S. 315 (1991).
o Id. at 319-20.
16, See Barry R. Goldman, Note, Can the King Do No Wrong? A New Look at the DiscretionaryFunc-
tion Exception to the Federal Tort Claims Act, 26 GA. L. REV. 837, 851-52 (1992).
'62 It is worth noting, however, that some observers have argued that Washington, the only
state without government-tort immunity, has experienced serious financial consequences as a
result of litigation costs. Tardif & McKenna, supra note 27, at 44-53.
"'5 See SCHUCK, supra note 106, at 60-66. See generally GORDON TULLOCK, THE POLITICS OF
BUREAUCRACY 157-77 (1965); Isaac Ehrlich & Richard A. Posner, An Economic Analysis of Legal
Rulemaking, 3J. LEGAL STUD. 257, 263 (1974).
'" See, e.g., WILLIAM A. NISKANEN,JR., BUREAUCRACY AND REPRESENTATIVE GOVERNMENT 138-
54 (1971).
'w Bureaucratic insulation from political control is likely to be reduced when elected officials
have a low-cost means of monitoring bureaucratic conduct. See generallyJeffrey S. Banks & Barry
R. Weingast, The Political Control of Bureaucracies Under Asymmetric Information, 36 AM. J. POL. SC.
509 (1992) (detailing the informational advantage bureaucrats possess over politicians and how
Feb. 2007] A THEORY OF GOVERNMENTAL DAMAGES LIABILIFY 837
The existence of term limits that prevent incumbents from seek-

ing reelection in some jurisdictions might also be expected to reduce
the concern of elected officials with the allocation of public resources
in order to enhance future electoral prospects, although the effect of
term limits is far from clear. Even officials barred from seeking re-
election may wish to use public resources to build support for elec-
tion to another office or for a designated successor; party leadership
may also seek to use pork-barrel policies in an effort to win or retain
an office that the incumbent must vacate as a result of term limits;
and by increasing the number of newly elected officials who may feel
especially vulnerable at the next election, term limits may actually in-
crease the number 66of incumbents particularly concerned about dis-
tributional politics.
A final objection to the theory advanced above comes from Pro-
fessor Levinson himself. We have seen that he believes that elected
officials actually welcome tort liability as a means for deflecting oppo-
sition to their policies. 67 If that is right, then governmental damages
liability actually yields political benefits.
Thus, in the absence of persuasive empirical evidence, it is fairly
debatable whether the theory of political behavior advanced above
has merit. Having observed municipal government at close range for
many years, I will confess that it has always been obvious to me that
elected officials would always prefer to control the disposition of pub-
lic funds rather than yield control to judges and juries, if for no other
reason than to be able to take credit when funds are paid out-even
when paid as compensation to those injured by governmental con-
duct. But surely this debate should be resolved on the basis of some-
thing other than my own anecdotal experiences. There is, however,
empirical evidence available to test the reaction of elected officials to
the threat of tort liability-the existence of tort immunity legislation.
politicians seek to mitigate such an advantage in their design of agencies). The fact that bu-
reaucrats must ask the legislature for an annual appropriation to pay judgments gives elected
officials a ready means of monitoring bureaucratic risk management, and therefore suggests
that political control over litigation costs is relatively high.
' For a discussion of the complicated effect of term limits on distributional politics, see Dan
Bernhardt et al., Term Limits and Pork Barrel Politics, 88J. PUB. ECON. 2383 (2004). For a discus-
sion of how, for politicians, removal of the reelection preference elevates the relative impor-
tance of wielding power, setting policy, and reaping benefits from special interests, see Eliza-
beth Garrett, Term Limitations and the Myth of the Citizen-Legislator, 81 CORNELL L. REV. 623
(1996). For empirical evidence that term limits actually increase electoral competition by re-
ducing the number of entrenched incumbents, see Kermit Daniel &John R. Lott, Jr., Term Lim-
its and Electoral Competitiveness: Evidence from California's State Legislative Races, 90 PUB. CHOICE
165 (1997).
...See supra text accompanying note 120.
3. Immunity Legislation as EmpiricalEvidence in Support of the Theory

We have seen that statutory tort immunity is ubiquitous; the fed-
eral government and forty-nine states have legislation offering gov-
ernmental defendants substantial protection from tort liability.' En-
acting such legislation, however, is not politically costless; it is
reasonable to expect political opposition to tort immunity. From the
standpoint of the economic interests of the typical voter, tort-
immunity legislation is a wash, since whatever risk of uncompensated
loss an individual voter is likely to run should be roughly equal to the
additional taxes that the voter must pay to finance government litiga-
tion costs. But voters might be expected to oppose tort immunity on
the basis of the widely held belief that people who are wrongfully in-
jured should be entitled to recover fair compensation from the
wrongdoer, a point that S ••corrective
169 justice theorists argue reflects a
deeply held moral intuition.
Public-choice theory predicts that the problems of collective ac-
tion, present when the stakes for each individual are relatively small,
mean that the holders of particularly large economic interests in pol-
icy questions will play a disproportionate role in the political proc-
ess.'C This view suggests that interest groups with a substantial finan-
cial interest in pressing tort liability will be a particularly important
source of political opposition to immunity. Indeed, the plaintiffs' bar
acts as a potent lobby opposing legislation restricting tort liability.''
On governmental immunity issues, it could form alliances with any
number of interest groups that have a financial interest in govern-
mental liability, such as trucking firms that experience losses when
public roads are not well designed or maintained, or businesses that
experience losses when negligent or otherwise unlawful governmen- I
tal licensing or other regulatory decisions curtail their operations. 1
See supra text accompanying notes 21-52.

'6 See supra text accompanying note 103.
"0 For a sampling of the leading public choice literature on this point, see, for example,
MANCUR OLSON, JR., THE LOGIC OF CoLLECTIvE ACTION: PUBLIC GOODS AND THE THEORY OF
GROUPS 111-67 (1965); Gary S. Becker, A Theory of CompetitionAmong Pressure Groupsfor Political
Influence, 98 Q.J. ECON. 371 (1983); Fred S. McChesney, Rent Extraction and Rent Creation in the
Economic Theory of Regulation, 16J. LEGAL STUD. 101 (1987); and GeorgeJ. Stigler, The Theory of
Economic Regulation, 2 BELLJ. ECON. & MGMT. SCI. 3 (1971).
171 See, e.g., Paul H. Rubin & Martin J. Bailey, The Role of Lawyers in Changingthe Law, 23 J.
LEGAL STUD. 807, 827 (1994).
17 Opposition to governmental-immunity legislation, among other things, would
also reflect
the demoralization costs likely to ensue if residents of a jurisdiction are required to run the risk
that they will suffer an injury at the hands of the government, or a government official for
which they can receive no compensation. The concept of demoralization costs was first ad-
vanced by Professor Michelman as one of the justifications for the prohibition on uncompen-
sated takings, Michelman, supra note 99, at 1214-18, but it has application to governmental tort
immunity as well, id. at 1169 n.5.
And on particular liability issues, any number of lobbies are likely to

press for governmental liability, such as community organizations
concerned about police brutality, domestic-violence victims' advo-
cates concerned about police indifference to this issue, bicyclists' ad-
vocates concerned about improving the design and maintenance of
bike paths, patients' rights groups concerned about the quality of ser-
vice provided by public ambulances and hospitals, and so on.
In contrast, other than the government itself, it is difficult to iden-
tify a lobby that should favor governmental tort immunity. The in-
surance industry might ordinarily be expected to advocate immunity,
but governmental immunity works against insurance interests by lim-
iting the ability of insurance companies to seek recovery from the
government when its wrongful acts cause insured losses to policy-
holders whose rights the insurers can assert through subrogation or
similar arrangements.13 Taxpayers' advocates and public-employee
unions might have some interest in this issue, but they would be far
more likely to pursue their interests directly-by lobbying for tax cuts
or indemnity, respectively-rather than deploying their limited re-
sources and political capital on an issue that only indirectly affects
their core interests. I know of no evidence that these groups-or any
other organized lobby aside from those controlled by elected offi-
cials-have ever taken a position on governmental tort-immunity leg-
islation. Indeed, when I was in municipal government, we frequently
tried to interest these groups in lobbying on governmental tort im-
munity issues; we never succeeded. But even if these groups were
thought to be vigorous proponents of immunity legislation, that
point would only strengthen the theory of political behavior ad-
vanced above. After all, to the extent that tort liability increases tax
burdens-thereby activating the taxpayers' lobby-or reduces the
government's ability to provide its employees with their desired com-
pensation-thereby activating public-employee unions-then tort li-
ability imposes a political price on elected officials.
In short, the ubiquity of governmental tort immunity legislation
itself suggests that tort immunity confers political benefits on incum-
bents. It therefore follows that tort liability exacts a political price.
For this reason, the claim that tort liability cannot be expected to
produce governmental investments in loss prevention should not
persuade.
73 Such arrangements are common in the insurance industry as a vehicle by which insurance
companies are able to sue those responsible for insured losses. See, e.g., Jeffrey A. Greenblatt,
Insurance and Subrogation: When the Pie Isn't Big Enough, Who Eats Last?, 64 U. CHI. L. REv. 1337,
1339-42 (1997).
4. CorroborativeEvidencefor the Theory
Thus, government-immunity statutes powerfully suggest that

elected officials are sensitive to tort liability. But additional evidence
can also be mustered in support of this conclusion. Consider the af-
termath of FloridaPrepaid PostsecondaryEducation Expense Board v. Col-
lege Savings Bank. 7 4 In that case, the Supreme Court held that federal
legislation subjecting states to damages liability for patent infringe-
ment violated the Eleventh Amendment's prohibition on damages
actions against nonconsenting states.'75 Given the strength of the
patent lobby and its interest in this issue as evinced by its ability to ob-
tain federal legislation that imposed liability on the states (but, sig-
nificantly, without a corresponding waiver of federal sovereign im-
176
munity), the patent lobby should have been able to secure state
legislation granting patentees the right to seek damages for patent in-
fringement. Yet such legislation has not been forthcoming. 177 It is
difficult to understand this apparent lobbying failure unless state leg-
islatures and governors perceive a cost to state patent infringement
liability that was not apparent to Congress, which saw no obstacle to
refusing the entreaties of the patent lobby when federal revenues
were not at risk.
The history of government-immunity legislation provides addi-
tional support for the theory advanced above. Prior to the enactment
of the FTCA, there was a long history of Congress granting legislative
relief to victims of federal torts until it found itself overwhelmed by
the volume of requests for legislative relief."8 That is itself a fairly
clear indication of the political momentum behind compensation.
Even since the FICA, this political dynamic has asserted itself; after
the Supreme Court, in Dalehite v. United States, 79 held that the United
States enjoyed discretionary immunity from liability for an explosion
and fire, 80 Congress nevertheless felt compelled to enact the Texas
City Disaster Relief Act, which authorized compensation while miti-
gating budgetary impact by capping permissible recovery at $25,000
per claimant."" This history suggests that there is political potency to
' 527 U.S. 627 (1999).

,75 Id. at 635-48.
"6 See 35 U.S.C. § 271(h) (2000 & Supp. I1 2003).
177 See, e.g., Jason Karasik, Note, Leveling the IP Playing Field: Conditional
Waiver Theory and the
Intellectual Property ProtectionRestorationAct, 27 HASTINGS COMM. & ENT. L.J. 475, 498 (2005).
.78 See, e.g., Floyd D. Shimomura, The History of Claims Against the United States: The Evolution
from a Legislative Toward a JudicialModel of Payment, 45 LA. L. REV. 625, 682-84 (1985).
79 346 U.S. 15 (1953).
,8 Id. at 26-32.
...Ch. 864, § 5(a), 69 Stat. 707, 708 (1955), amended by 1959 Amendment to the Texas City
Disaster Relief Act, Pub. L. No. 86-381, 73 Stat. 706 (providing additional relief). The statute
also did not compensate losses that had been indemnified by insurance, § 6(c), and limited
Feb. 2007] A THEORY OF GOVERNMENTAL DAMAGES LIABILFTY
the demand that compensation be provided to the victims of gov-

ernmental torts. There must be some offsetting political benefit to
explain why elected officials enact immunity statutes that deny their
constituents compensation.
Evidence that tort liability exacts a political price by reducing po-
litical control over public resources is also reflected in the pattern of
immunity legislation. If the relevant political calculus was a product
solely of the strength of the various lobbies pressing their position on
a legislature, then one would expect the pattern of statutory immu-
nity to reflect the strength of the various lobbies. 8 2 Some of that is
going on in immunity legislation; the statutes that immunize gov-
ernments from paying damages to prisoners, for example, seem to re-
flect a legislative loss by a disfavored group lacking political influence.
But the most common categorical immunity, for discretionary deci-
sions, is aimed at no particular class of plaintiffs, but instead is di- 3
rectly aimed at protecting the political prerogative to set policy.
The same is true of the even more common caps on damages recov-
erable from governmental defendants. 8 4 This pattern suggests that
the interest in maintaining control over public resources drives the
enactment of immunity legislation rather than the relative strength of
competing lobbies.
But perhaps all this is beside the point. The existence of immu-
nity legislation is itself the strongest evidence for the conclusion that
elected officials are highly sensitive to governmental damages liabil-
ity. If, as predicted by Levinson, elected officials were indifferent to
liability, they would not bother to enact immunity legislation. Even
Congress, armed with the resources of the iassive federal budget,
has been unwilling to subject the federal government to unlimited
tort liability. The prevalence of immunity legislation powerfully sug-
gests that tort immunity confers a political advantage. It follows that
tort liability itself must impose some correlative political cost such
that elected officials perceive a political advantage in the enactment
of immunity legislation.
permissible attorney's fees, § 11. The federal government ultimately paid $17.1 million in
claims on total losses estimated from $300 million to several billion dollars. COMM'N ON GOV'T
PROCUREMENT, 4 REPORT OF THE COMMISSION ON GOVERNMENT PROCUREMENT 99 (1972).
"2 This is the view of the legislative process ordinarily taken by public-choice theorists, for
example, who see political decision-making as a consequence of the interaction of the lobbies
with particularly significant stakes in various policy issues. See, e.g., DANIEL A. FARBER & PHILIP P.
FRICKEY, LAW AND PUBLIC CHOICE: A CRITICAL INTRODUCTION 12-21 (1991).
8' SeeKrent, supra note 57, at 1545-51; Niles, supra note 53, at 1301-05.
184 See supra note 49.
C. The Impact of Governmental Tort Liability

Thus, Professor Levinson is wrong to believe that governmental li-
ability has only indeterminate effects. Whatever its defects from the
standpoint of corrective justice, governmental tort liability has an in-
strumental justification; it creates an incentive on the part of office-
holders to allocate resources to loss prevention. There should be a
clear political incentive to invest in loss prevention at least when the
cost of avoiding an injury is small, the likelihood of injury is great,
and the impact on the government's budget is likely to be large.
Similarly, the Takings Clause's compensation requirement functions
as a political restraint on the use of the power of eminent domain.
Government will not take property unless the political benefits of the
taking exceed the political cost of compensation.
But under the view of political behavior advanced here, Professor
Levinson was right to claim that governmental tort liability has no ef-
ficiency justification comparable to the role of tort liability in the pri-
vate sector. In the private sector, tort liability creates an incentive to
invest in loss prevention in order to avoid an expected greater liabil-
ity, discounted to present value. If elected officials were concerned
only with maximizing the public resources available to be allocated
consistent with their own political preferences, then public officials
would be subject to the same incentive. But surely Levinson is right
that voters do not use any firm value equivalent for judging the op-
eration of government, and whenever the government chooses to in-
vest in loss prevention, it will experience a political opportunity cost
associated with the political benefits that some alternative use of
funds would have yielded. Therefore, there is always a political cost
to making investments in safety; some other use of the funds, and its
attendant political benefits, must be foregone, unless taxes are raised
or, if possible, additional debt is incurred, and those options carry po-
litical costs as well. If, for example, elected officials invest greater
funds in the internal-affairs division of the police department in or-
der to reduce police-related liability, they will have less money avail-
able to hire additional officers, unless they are willing to raise taxes
and therefore incur an additional set of political costs. Accordingly,
there will be substantial political opportunity costs when elected offi-
cials forego the opportunity to put more police on the streets, even
when they do so for the sake of risk management. The political op-
portunity cost of spending additional funds on internal affairs may
well be perceived to be higher than the political cost of failing to re-
duce future police-related liability; without knowing a great many
things about the political facts on the ground, it is impossible to make
confident predictions on this score.
Even aside from political opportunity costs, liability-producing
conduct may have political benefits that offset the deterrent effect of
liability. To use Professor Levinson's example, a program of aggres-

sive stop-and-frisk of young males in high-crime areas may increase
liability, but it also may pay such handsome political benefits that li-
ability will have no deterrent effect. 85 We can expect deterrence only
when the political cost of losing control over litigation-related costs
outweighs the political benefits of the program, and that calculation
is necessarily indeterminate. Similarly, we can expect exercise of
eminent-domain powers as long as the political opportunity costs of
paying just compensation do not exceed the political benefits of the
condemnation. There is, however, no ready way to monetize political
costs and benefits. Thus, governmental tort liability cannot guaran-
tee an "efficient" level of torts, constitutional torts, or takings.
This observation is not meant to be an argument against govern-
mental damages liability. The political costs attendant to governmen-
tal liability will operate as a political restraint on governmental torts
and takings that exceeds that present in a nonliability regime. At the
same time, there is little reason to believe that the government will
overinvest in loss prevention, or underinvest in eminent domain.
Tort liability gives elected officials an incentive to make only those in-
vestments that will likely produce even greater cost savings through
reduced liability, but no more. There is little reason to believe that
elected officials will reduce their own pool of funds available for po-
litically optimal use by overspending on loss prevention; the political
benefits of those investments are speculative, and most of them will
likely be realized at some uncertain point in the future, which may
well be beyond the next electoral cycle. 6 As for takings, as long as
the political benefits of a taking exceed the political costs of compen-
sation, condemnation will occur, and the most socially desirable con-
demnations are likely to pay the greatest political benefits. Thus, a
regime of governmental liability is likely to fail to achieve the level of
efficiency thought possible in the private sector, but it will lead to
greater investment in loss prevention than will a nonliability regime,
without creating any significant risk of overdeterrence.
" See supra text accompanying notes 114-15.

"6 Indeed, there is reason to believe that, because of the difficulty in estimating potential
tort liability ex ante, tort liability generally produces suboptimal investments in safety even in
the private sector. See, e.g., John E. Calfee & Richard Craswell, Some Effects of Uncertainty on Com-
pliance with Legal Standards, 70 V. L. REV. 965, 976-84 (1984); Richard Craswell, Deterrence and
Damages: The MultiplierPrinciple and Its Alternatives, 97 MICH. L. REV. 2185, 2191-98 (1999). This
is not to say that government never overinvests in loss prevention. There is increasing evidence
that individuals tend to be overly concerned with certain types of relatively small risks. See gener-
ally Timur Kuran & Cass R. Sunstein, Availability Cascades and Risk Regulation, 51 STAN. L. REV.
683, 685-89 (1999); Roger G. Noll & James E. Krier, Some Implications of Cognitive Psychology for
Risk Regulation, 19J. LEGAL STUD. 747, 747-48 (1990). When elected officials respond to these
concerns by overinvesting in loss prevention, however, they are responding to political pressure
and not the threat of tort liability.
III. ASSESSING GOVERNMENTAL LIABILITY
Parts I and II endeavor to describe the contours and operation of

governmental tort liability. It remains to offer an assessment of a re-
gime of governmental damages liability.
We have seen that governmental tort liability can be expected to
encourage elected officials to invest in loss prevention with little like-
lihood of overdeterrence. That may tempt one to favor governmen-
tal liability as a means of reducing welfare losses caused by govern-
mental conduct. Indeed, the law reviews are full of proposals for
damages claims that can be brought against the government to re-
duce the incidence of asserted welfare losses that are either attribut-
able to government or avoidable through greater governmental in-
vestment in loss prevention. Recent proposals would require the
government to pay damages to pretrial detainees who are ultimately
acquitted, 8 7 those whom the police fail to protect from mob vio- 189
lence, 11 8 students who are the victims of pregnancy discrimination
or obesity harassment,'90 victims of racial profiling,' 9' and those who
are wrongfully convicted.' 92 Scholars have also advocated the aboli-
tion of qualified immunity on the grounds that it too often leaves
constitutional violations unremedied,"' the abolition of absolute
prosecutorial immunity for the same reason, 94 imposing vicarious li-
ability on municipalities for the constitutional torts of their employ-
ees,1% and permitting awards of presumed and punitive damages
1"7 See Jeffrey Manns, Liberty Takings: A Framework for Compensating Pretrial Detainees, 26
CARDOZO L. REV. 1947 (2005).
"' See Susan S. Kuo, Bringing in the State: Toward a ConstitutionalDuty to Protect from Mob Vio-
lence, 79 IND. L.J. 177 (2004).
See David S. Cohen, Title X: Beyond Equal Protection, 28 HARv.J.L. & GENDER 217 (2005).
SeeJessica Meyer, Obesity Harassment in School: Simply "Teasing" Our Way to Unfettered Obesity
Discriminationand StrippingAway the Right to Education,23 LAw & INEQ. 429 (2005).
'9'See David Rudovsky, Law Enforcement by Stereotypes and Serendipity: Racial Profilingand Stops
and Searches Without Cause,3 U. PA.J. CONST. L. 296 (2001).
2 See Brandon L. Garrett, Innocence, Harmless Error, and Federal Wrongful Conviction Law, 2005
WIS. L. REV. 35; Michael Goldsmith, Reforming the Civil Rights Act of 1871: The Problem of Police
Perjury, 80 NOTRE DAME L. REV. 1259 (2005).
93 See, e.g., Mark R. Brown, Weathering Constitutional Change, 2000 U. ILL. L. REV. 1091, 1101-
10; Alan K. Chen, The Ultimate Standard: Qualified Immunity in the Age of ConstitutionalBalancing
Tests, 81 IOWA L. REV. 261, 307-33 (1995); Pillard, supra note 89, at 91-103; Daniel L. Roten-
berg, Private Remedies for Constitutional Wrongs-A Matter of Perspective, Priority, and Process, 14
HASTINGS CONST. L.Q. 77, 90 (1986); Rudovsky, supra note 143, at 1217-26; Christina B. Whit-
man, Government Responsibilityfor ConstitutionalTorts, 85 MICH. L. REV. 225, 257-76 (1986).
' See Margaret Z. Johns, ReconsideringAbsolute ProsecutorialImmunity, 2005 BYU L. REV. 53.
'95See, e.g.,
SHUCK, supra note 106, at 82-121; Jack M. Beermann, Municipal Responsibilityfor
ConstitutionalTorts, 48 DEPAUL L. REV. 627 (1999); Mark R. Brown, CorrelatingMunicipalLiability
and Official Immunity Under Section 1983, 1989 U. ILL. L. REV. 625, 669-88; Kramer & Sykes, supra
note 111, at 272-96; Harold S. Lewis, Jr., & Theodore Y. Blumoff, Reshaping Section 1983s Asym-
metry, 140 U. PA. L. REV. 755, 825-48 (1992); Laura Oren, Immunity and Accountability in Civil
Rights Litigation: Who Should Pay?, 50 U. PIr. L. REV. 935, 1000-07 (1989).
Feb. 2007] A THEORY OF GOVERNMENTAL DAMAGES LIABILFIT 845
against municipalities in order to maximize the deterrent effect of

civil liability.'"6 Each of these proposals requires the government to
assume costs in order to avoid losses experienced by others; conse-
quently, they all have a negative impact on government budgets, re-
gardless of the externalized benefits they may produce-unless one
can make the rather implausible claim that these proposals would be
so popular that the voters would tolerate an increase in taxes to fund
the new expenditures that they necessitate. Yet one cannot find in
any of the proposals for new governmental liability any consideration
of the consequences that new liabilities will have on government
budgeting, or on those who depend on government budgets for the
variety of social goods allocated through that process.
In fact, governmental liability has important effects on govern-
ment budgeting that are likely to be concentrated among the most
disadvantaged in society. Gerald Frug once remarked on this phe-
nomenon when considering court-ordered remedial plans for public
institutions thought to be performing below constitutional standards:
Because government resources are limited and because some commit-
ments of those resources cannot be reduced due to contract or other ob-
ligations, the impact of a court's decision falls on a relatively few budget
items. The court is in fact allocating the budget away from those items,
probably without even knowing what they are. The court's allocation de-
cision is simply that every element of the court decree take precedence
over every other competing element in the budget, whatever they may be.
The legislature retains no say at all about the comparative value of the
item lost to the item required by the court. Thus the value of legislative
decisionmaking on budget allocation is undermined, to a greater or
lesser degree, depending on the size of the court's demands and the
amount of money available.
Some have argued that such judicial intervention in the budget proc-
ess in favor of prisoners and the mentally ill can be justified because
those groups are left out of the normal political decisionmaking proc-
esses. Indeed they often are. But the scarce resources allocated by gov-
ernment are largely allocated to people indistinguishable from those af-
fected by the court orders. The mentally ill involuntarily committed to
an institution may receive additional services under court order at the
expense of those voluntarily committed to the same institution, or those
not committed but using outpatient facilities at public hospitals or men-
tal health centers. Prisoners may receive better medical care at the ex-
pense of a parolee who seeks it at a public hospital, or they may receive
training or addiction services at the expense of the public at large who
need identical services. Many beneficiaries of court orders are not enti-
tled to vote, but neither are the children whose access to education, li-
braries, or welfare benefits might be curtailed to pay for the court order.
See Rudovsky, supra note 143, at 1225.

JOURNAL OF CONSTITUIMONAL LA W [Vol. 9:3
The allocation of scarce resources by court order is not likely to be fortu-

nate to the powerless;97 it is already the powerless to whom the state largely
directs its resources.1
Professor Frug's point has even more force when applied to dam-
ages awards against the government-damages awards drain the pub-
lic treasury in an even more direct fashion than compliance with in-
junctive decrees. What is more, it is far from clear that social welfare
is enhanced by a regime of governmental tort liability. If one as-
sumes that the public's willingness to pay taxes is essentially fixed at
any given time, then diverting limited revenues from public uses that
may produce greater social welfare gains-infrastructure, education,
health care, police, and the host of other welfare-enhancing services
that government provides-to the payment of judgments and other
litigation expenses could well harm social welfare. After all, the exis-
tence of political accountability creates incentives on the part of
elected officials to spend public funds in ways that produce relatively
broad benefits that will yield a winning coalition at the next elec-
tion-an incentive that tort plaintiffs and their lawyers lack. Thus,
even if one is willing to ignore the distinction between insurance and
tort and advocate governmental liability merely on grounds of loss-
spreading, it is far from clear that social welfare will be advanced if
government resources must be diverted from other uses to what is es-
sentially the provision of publicly funded tort (or takings) insurance,
which, of course, can otherwise be purchased in a private market by
those concerned about the risk of uncompensated loss.
Finally, and relatedly, governmental tort liability also has a signifi-
cant impact on two especially large groups of third parties-the tax-
payers and the public at large. We have seen that the taxpayers lack
meaningful culpability for the tortious conduct of government and
are not similarly situated to the owners of a private-sector entity who
must shoulder the costs of liability, yet the financial consequences of
governmental liability are borne by the taxpayers nevertheless.
Moreover, whenever a judge or jury unilaterally directs a commit-
ment of government resources to a particular plaintiff, requiring the
imposition of additional taxes or a reordering of budgetary priorities,
republican values are compromised. These budgetary effects can
have dramatic impacts on those who are dependent on government
197 Gerald E. Frug, TheJudicialPower of the Purse, 126

U. PA. L. REV. 715, 741-42 (1978) (foot-
notes omitted). As far as I am aware, the rich literature on institutional reform litigation con-
tains no answer to Professor Frug on this point. For a sampling of the literature and its explica-
tion of what are seen as the virtues ofjudicial management of public institutions, see generally
Abram Chayes, The Role of the Judge in Public Law Litigation, 89 HARV. L. REV. 1281 (1976); Owen
M. Fiss, The Supreme Court, 1978 Term-Foreword: The Forms ofJustice, 93 HARV. L. REV. 1 (1979);
William A. Fletcher, The DiscretionaryConstitution: InstitutionalRemedies andJudicialLegitimacy, 91
YALE L.J. 635 (1982); and Paul Gewirtz, Remedies and Resistance, 92 YALE L.J. 585 (1983).
services. After all, government is subject to an array of responsibili-

ties unknown in the private sector. While a private party can rea-
sonably be expected to make all cost-justified investments in safety,
private tortfeasors are generally under no duty to protect the public
at large from threats not of their own creation.'" Government, how-
ever, has a politically enforceable obligation to protect the public
from all threats to its safety and welfare. Accordingly, government
faces infinitely more difficult resources-allocation decisions than
those confronting the private sector. When ajudge or jury exercises
control over the allocation of public resources, the consequences may
be far more dramatic than in the private sector if the ability of the
government to meet the social, economic, security, and other threats
the public expects it to confront is compromised by the burden of
litigation costs.
Thus, governmental damages liability is problematic, to say the
least. With that preface, the various species of governmental liability
merit separate consideration.
A. Common-Law Torts
1. The Marginal Utility of Governmental Liability

We have seen that governmental tort liability can be expected to
produce greater governmental investment in loss prevention than a
nonliability regime. There is reason to doubt, however, that the mar-
ginal gain in loss prevention will be significant.
Aside from the system of tort liability, there is another means for
holding the government accountable when it fails to make sufficient
efforts to prevent losses fairly attributable to its own activities-the
next election. When the government fails to invest sufficient re-
sources in the maintenance of its streets, for example, the issue can
arise in the next political campaign regardless of whether vehicle
owners are able to bring tort actions to recover damages caused by
potholes. Indeed, the political potency of this issue may well be
greater if the vehicle owners' losses have gone uncompensated. The
existence of such political accountability in government calls into
question any regime of common-law tort liability. Further, at least at
the local level, there is an especially stringent form of political ac-
countability at work. Because at least some businesses and individuals
are able to opt out of the local political process by moving to a differ-
ent location, and because new businesses and residents are always
free to select their desired location, local governments are effectively
t" See RESTATEMENT (SECOND) OF TORTS §§ 314-15 (1965).

JOURNAL OF CONSTITUFTIONAL LA W [Vol. 9:3
competing with each other for desirable residents and businesses that
can assist in local development.'" Considerable empirical evidence
confirms that local governments compete to provide residents and
businesses with a high ratio of services to taxes as a result of these
competitive pressures.0 0 It follows that, even absent governmental
tort liability, if a local government failed to make adequate invest-
ments in loss prevention, driving up local insurance rates and increas-
ing the risk of uncompensated loss, then residents and businesses will
locate elsewhere, undermining the local tax base and economy. This
dynamic creates powerful political incentives to invest in loss preven-
tion that do not depend on the threat of tort liability. Indeed, there
is some reason to believe that, even among the states, interjurisdic-
tional competition operates in this same fashion, albeit to a lesser ex-
tent.201
Accordingly, political accountability gives elected officials an in-
centive to protect the public's safety-their political vulnerability in
such circumstances can be at least as important as the threat of dam-
ages liability. To make the point concrete, consider once again Dale-
hite. That case was one of some three hundred arising from the ex-
plosion and fire at an ammonium nitrate facility at Texas City, Texas,
on A ril 16 and 17, 1947, seeking in total some $300 million in dam-
ages. As part of a federal program in which facilities that formerly
were used to make explosives were converted to the production of
fertilizer for occupied Japan, nearly three-thousand tons of ammo-
nium nitrate-based fertilizer were shipped to a warehouse in Texas
201
City and loaded onto two ships.• One of the ships
,,204caught fire, both
exploded, and "much of the city was leveled .... Justice Jackson
recounted the fallout from the disaster:
More than 560 persons perished in this holocaust, and some 3,000 were
injured. The entire dock area of a thriving port was leveled and property
damage ran into the millions of dollars.
This was a man-made disaster; it was in no sense an "act of God." The
fertilizer had been manufactured in government-owned plants at the
Government's order and to its specifications. It was being shipped at its
direction as part of its program of foreign aid. The disaster was caused
by forces set in motion by the Government, completely controlled or
'9 See, e.g., NISKANEN, supra note 164, at 155; PETERSON, supra note 149, at 17-38, 71-77;
Charles M. Tiebout, A Pure Theory ofLocal Expenditures, 64J. POL. ECON. 416, 418 (1956).
m See WILLIAM A. FISCHEL, REGULATORY TAKINGS: LAw, ECONOMICS, AND POLITICS 254-69
(1995); Vicki Been, "Exit" as a Constraint on Land Use Exactions: Rethinking the Unconstitutional
Conditions Doctrine,91 COLUM. L. REV. 473, 512-28 (1991).
20' See Barry R. Weingast, The Economic Role of PoliticalInstitutions: Market-PreservingFederalism
and Economic Development, 11 J.L. ECON. & ORG. 1, 5 (1995).

202 Dalehite v. United States, 346 U.S. 15, 17 (1953).
Id. at 18-23.
20I Id. at 23.
controllable by it. Its causative factors were far beyond the control or
knowledge of the victims; they were not only incapable of contributing to
it, but could not even take shelter or flight from it. 205
The majority concluded that the allegations that the government
negligently planned and executed the shipment of unstable fertilizer
were barred by the FTCA's immunity for discretionary judgments.' °6
In dissent, Justice Jackson argued that the government officials run-
ning the program should have foreseen that the fertilizer was unsta-
ble and posed an unreasonable threat to the public. °7 Justice Jack-
son's dissent makes a powerful argument that government officials
had recklessly endangered the public-an argument with enormous
political potency, whatever its legal merit. Indeed, we have seen that
Congress eventually provided statutory compensation for the victims
of the Texas City disaster, albeit at a level with less budgetary impact
than an award of complete consequential damages.2 0 s As the events
surrounding Dalehite illustrate, political forces come powerfully into
play when the government endangers the public's safety, even when
there is immunity from liability.
Consider another example of more recent vintage. It has been
reported that the failures in New Orleans' levee system caused by
Hurricane Katrina were the result of easily foreseeable and readily
repairable faults in levee construction and maintenance. 09 That may
make out a classic case of negligence,2 0 but surely no one could think
that such a damages action would be necessary to hold public officials
accountable for their failure to maintain the levee system properly."'
Id. at 48 (Jackson,J., dissenting).

Id. at 35-43 (majority opinion).
o7 Id. at 50-53 (Jackson, J., dissenting).
' See supra text accompanying note 181. The legislative history suggests that the primary
motive for the relief act was a congressional acknowledgement that the Texas City disaster had
been the consequence of government negligence. See H.R. Rep. No. 83-1386, at 32-33 (1954);
S. Rep. No. 83-2363, at 3 (1954).
m See Bob Marshall, Doomed to Failure: State Report Says Bad Design Led to 17th Street Canal
Breach, TIMES-PICAYUNE (New Orleans, La.), Nov. 30, 2005, at A-1; John Schwartz, Army Builders
Accept Blame over Flooding, N.Y. TIMES, June 2, 2006, at Al; John Schwartz, New Study of Levees
FaultsDesign and Construction,N.Y. TIMES, May 22, 2006, at A19; Bill Walsh, Corps ChiefAdmits to
Design Failure,TIMES-PICAYUNE (New Orleans, La.), Apr. 6, 2006, at A-1.
210 See Bruce Eggler, Lawoyer: Sue Feds on Levee; NO. Council Responds Eagerly, TIMES-PICAYUNE
(New Orleans, La.), Dec. 2, 2005, at B-1. It appears, however, that the federal government en-
joys statutory immunity from liability for water damage caused by levee failures. See 33 U.S.C.
§ 702c (2000) ("No liability of any kind shall attach to or rest upon the United States for any
damage from or by floods or flood waters at any place .... "); see also United States v. James, 478
U.S. 597, 608 (1986) ("Congress clearly sought to ensure beyond doubt that sovereign immu-
nity would protect the Government from 'any' liability associated with flood control."), overruled
on other grounds by Cent. Green Co. v. United States, 531 U.S. 425 (2001).
2 The flooding in New Orleans after Hurricane Katrina has provoked a flurry of accusations
in an effort to hold those thought to be responsible politically accountable. See, e.g., Frank
Donze, Sea of Levee Board Records Sought: Senate Probe Wants Lengthy PaperTrail, TIMES-PICAYUNE
JOURNAL OFCONSTITITIONAL LAW [Vol. 9:3
Any instance of government bungling that compromises the public's

safety is likely to have potent political consequences.
2. The Difficulty ofAdjudicating GovernmentalLiability

Then there is the problematic nature of adjudicating governmen-
tal liability in tort, as LouisJaffe observed decades ago:
We can assume the hypothesis of a governmental decision which creates
a risk that could be considered either unnecessary to achieve the end in
view or avoidable by an expenditure deemed reasonable. The difficulties
of evaluating such a decision in terms of negligence are notorious. The
decision will have rested in part on a technical judgment as to the size of
the risk and the need to incur it; in part on a political judgment as to
who should bear the indirect costs. Ajudge or ajury is not well equipped
either to make such determinations initially or to review them.212
For these reasons, ProfessorJaffee defended immunity for discretion-
ary decisions, "if only because, as we have noted, a court cannot un-
dertake to determine whether complex government decisions are
'reasonable.' 213 Indeed, this is the standard account of discretionary
immunity.214 But Professor Jaffee understated the problem; only in
rare tort actions is an evaluation of the manner in which the govern-
ment has chosen to allocate scarce public resources unnecessary.
Take as prosaic an example as the repair of potholes. 5 State and lo-
cal governments must decide how much money they will allocate to
(New Orleans, La.), Nov. 30, 2005, at A-i; James Gill, Op-Ed, Help! Corps Back to the Drawing
Board,TIMES-PICAYUNE (New Orleans, La.), Nov. 9, 2005, at B-7; Jeffrey Meitrodt, Doubts on Lev-
ees Cloud Conference: 40-Mile Fix May Leave Rest of System at Risk, TIMES-PICAYUNE (New Orleans,
La.), Nov. 11, 2005, at A-1; Mark Shleifstein, Levee Team Runs into Wall: It Reports No Access To Key
Records, Staff TIMES-PICAYUNE (New Orleans, La.), Oct. 26, 2005, at A-1.
212 Louis L. Jaffe, Suits Against Governments and Officers: Damage
Actions, 77 HARV. L. REv. 209,
235 (1963).
213 Id. at 237.
211 See, e.g., United States v. Gaubert, 499 U.S. 315, 322-23 (1991); Berkovitz v. United States,
486 U.S. 531, 536-37 (1988); United States v. S.A. Empresa de Viacao Aerea Rio Grandenese,
467 U.S. 797, 813-14 (1984). Considerations of this nature also underlie what many jurisdic-
tions have dubbed the public-duty doctrine, which provides that the government does not have
a duty in tort to provide government services, although the doctrine provides no defense to
what is thought to be affirmative governmental negligence or misconduct. See DAN B. DOBBS,
THE LAw OFTORTS §§ 271-72 (2000).
2"5 Litigation against government over an alleged failure to report potholes or other defects
in public roads is commonplace in the law of governmental torts. See, e.g., Nishihama v. City &
County of S.F., 112 Cal. Rptr. 2d 861, 863 (Cal. Ct. App. 2001); City & County of Denver v. Gon-
zales, 17 P.3d 137, 140 (Colo. 2001) (en banc); Steele v. Town of Stonington, 622 A.2d 551, 555
(Conn. 1993); Hanley v. City of Chi., 795 N.E.2d 808, 816 (Ill. App. Ct. 2003); Cooper v. La.
State Dep't of Transp. & Dev., 885 So. 2d 1211, 1219 (La. Ct. App. 2004); Minder v. Anoka
County, 677 N.W.2d 479, 487 (Minn. Ct. App. 2004); Townsend v. State, 738 P.2d 1274, 1277
(Mont. 1987); Chatman v. Hall, 608 A.2d 263, 276 (N.J. 1992), superseded by statute, Act of June
23, 1994, Ch. 49, 1994 N.J. Laws 322 (codified at NJ. STAT. ANN. § 59:3-1 (West 1992)); Lippel v.
City of N.Y., 722 N.Y.S.2d 511, 512-13 (N.Y. App. Div. 2001).
Feb. 2007] A THEORY OF GOVERNMENTAL DAMAGES LIABILITY
this function. When sufficient funds are not allocated to locate and
repair all potentially dangerous potholes as soon as practicable, the
government imposes upon all who use its roads a risk of injury. Still,
this is at its heart a discretionary judgment-government must decide
how much it can afford to spend on pothole repair in light of its lim-
ited resources and the many demands on those resources. The gov-
ernment may be willing to tolerate a higher risk of injury from pot-
holes rather than reduce the resources available for police
protection, which may lead to even more serious losses. For these
reasons, pothole repair can be characterized as falling within discre-
tionary immunity. 16 Even the method that repair crews use involves a
discretionary judgment-they might repair more patches more dura-
bly with more labor-intensive methods, but the more time that is
spent on each patch, the fewer potholes that can be repaired. 7
Accordingly, the line between immunized discretion and nonim-
munized negligence is wholly indistinct. To illustrate the point, con-
sider Indian Towing Co. v. United States.1 In that case, the Supreme
Court held that, although the Coast Guard's decision to provide a
lighthouse in aid of navigation was an immunized act of discretion,
once it undertook to provide that service and induced reliance on the
part of passing ships, it was obligated to keep the lighthouse in good
repair. But programs to inspect and repair lighthouses cost
money-the more frequent and thorough the inspections, the
greater the cost. The Coast Guard must decide how to undertake in-
spection and repair in light of its limited resources and the many
demands placed upon it. It is difficult to understand how a jury
could define the Coast Guard's duty of care with respect to its light-
houses without revisiting a host of policy decisions made by Congress
in its appropriations and by Coast Guard officials in allocating funds
within appropriated limits. Perhaps the Coast Guard abuses its dis-
cretion by failing to promptly inspect and maintain lighthouses-
although even this much is debatable inasmuch as committing re-
sources in this fashion could leave even more serious hazards to navi-
gation unaddressed-but its resource-allocation decisions, even if
misguided, are an immunized discretionary function nevertheless.
Thus, although there are perhaps a few government decisions that do
not involve any meaningful consideration of how limited resources
216 See Minder, 677 N.W.2d at 487; Mitchell v. City of Trenton, 394 A.2d 886, 888 (N.J. Super.
Ct. App. Div. 1978).
2,7 See Wrobel v. City of Chi., 742 N.E.2d 401, 406 (Ill. App. Ct. 2000).
2" 350 U.S. 61 (1955).
219 Id. at 69.
JOURNAL OF CONSTFTUTIONAL LAW [Vol. 9:3
are to be allocated, 220 any time that a plaintiff contends that a gov-
ernmental defendant failed to undertake sufficient investment in loss
prevention, it is entirely fair to characterize
22 the lawsuit as seeking to
impose liability on a policyjudgment. '
The problem of discretion, in turn, illustrates the anomalous role
that courts must assume in assessing governmental tort liability. Most
government decisions-how much time to spend training police offi-
cers in the use of excessive force, whether and how to investigate al-
legations of misconduct by public officials, whether to barricade pot-
holes in public streets-involve questions about how to allocate
scarce public resources. We have seen that the resources-allocation
decisions that the government faces are infinitely more difficult than
those confronting the private sector. Yet there is little reason to be-
lieve that judges or juries are in any position to pass judgment on
those decisions, since such judgments, after all, require consideration
of all the demands facing a unit of government with a responsibility
to protect the safety and welfare of the entire public. The Appropria-
tions Clause and its state counterparts suggest that it is a uniquely leg-
islative function to decide how public funds are to be allocated.2M Yet
judges and juries effectively exercise that function when they impose
tort liability on government.
To be sure, one can argue that tort law should require the gov-
ernment to raise taxes to the point where sufficient funds are spent to
provide reasonable protection from all threats, but the threat to re-
publican values would be profound if tort law, rather than the peo-
ple's elected representatives, took control of tax and loss-prevention
policy. Moreover, a decision to levy taxes at the level sufficient to
make all investments in loss prevention that a jury might find to be
costjustified would itself raise a host of complex issues-for instance,
whether increased taxes may adversely affect the business climate or
0 An example that the Supreme Court offered of negligence without any meaningful rela-
tion to the formulation of public policy or the allocation of public resources is negligent driving
by bank regulatory officials. See United States v. Gaubert, 499 U.S. 315, 325 n.7 (1991).
' It is therefore small wonder that commentators complain that the scope of discretionary
immunity is unclear and that it is applied inconsistently. See, e.g., John W. Bagby & Gary L. Git-
tings, The Elusive Discretionary Function Exception from Government Tort Liability: The Narrowing
Scope of FederalLiability, 30 AM. Bus. LJ. 223, 225 (1992); David S. Fishback & Gail Killefer, The
Discretionary Function Exception to the Federal Tort Claims Act: Dalehite to Varig to Berkovitz, 25
IDAHO L. REV. 291, 303 (1988-89); William P. Kratzke, The Supreme Court's Recent Overhaul of the
DiscretionaryFunction Exception to the Federal Tort Claims Act, 7 ADMIN. L.J. A. U. 1, 2-5 (1993);
Bruce A. Peterson & Mark E. Van Der Wiede, Susceptible to Faulty Analysis: United States v.
Gaubert and the Resurrection of Federal Sovereign Immunity, 72 NOTRE DAME L. REV. 447, 448
(1997).
" See supra text accompanying notes 14-20. For a similar argument made in the context of
an asserted right to receive government services, see Barbara E. Armacost, Affirmative Duties,
Systemic Harms, and the Due Process Clause,94 MICH. L. REV. 982 (1996).
otherwise reduce the attractiveness of the jurisdiction to the point

that the tax base will itself erode, jeopardizing the ability of the juris-
diction to afford continued investment in loss protection. And in
light of the political constraints on increasing taxes, governmental
tort liability could have the perverse result of reducing rather than
increasing governmental investment in loss prevention. Resources
diverted to litigation costs are unavailable for investment in loss pre-
vention, and, especially in a poorer jurisdiction with a limited tax
base, it may be impracticable to increase taxes sufficiently to cover
both litigation costs and all other measures that a jury might one day
conclude constituted cost-justified investment in loss prevention.
Moreover, the financial burden of liability falls on taxpayers who, as
we have seen, are essentially innocent third parties with far less ability
to "exit" a taxing jurisdiction than a shareholder dissatisfied with cor-
porate risk management has to sell his stock.
These problems have a racial dimension as well. As Richard
Thompson Ford has explained, even in a race-neutral legal regime,
racial and ethnic groups that have faced discrimination in the past
are disproportionately likely to be poor and therefore to reside in
older, less desirable communities where housing is relatively inexpen-
sive.2 3 These communities will, in turn, more likely experience
crumbling infrastructure and a host of other problems that give rise
to tort liability, yet their tax bases will make it all the more difficult to
finance necessary improvements. Moreover, when the government
must use tax revenues to pay judgments and other legal expenses that
fall disproportionately on such communities, businesses and high-
income taxpayers will experience a declining ratio of government ser-
vices received to taxes paid and will have a greater incentive to locate
elsewhere. The remaining tax base is less able to afford to pay tort
judgments and to fund the necessary cost-avoidance measures.
Thus, tort liability is a kind of regressive tax likely to impose
greater burdens on poorer communities with lesser ability to fund ei-
ther the liabilities or the necessary loss-prevention measures. It is ac-
cordingly one of a number of factors that can stimulate urban decline
and exacerbate problems of racial unfairness.
3. The Weak Casefor Governmental Liability
There is accordingly a strong argument to be made that govern-

ment loss-prevention policies are best left to the ordinary process of
political accountability by which other government policies are as-
sessed, rather than effectively controlled through the vehicle of tort
See Richard Thompson Ford, The Boundaries of Race: Political Geography in Legal Analysis,
107 HARv. L. REv. 1841, 1849-52 (1994).
JOURNAL OF CONSTFITUTIONAL LAW [Vol. 9:3
litigation. Nevertheless, arguments for governmental tort liability

remain.
Political accountability is at best an imperfect means for encourag-
ing government to undertake adequate investment in loss prevention.
The advocates of public-choice theory teach that political account-
ability operates imperfectly because it is generally costly and difficult
for the voters to monitor government policy.224 Surely this observa-
tion has particular merit when it comes to investments in loss preven-
tion; it is not very often that government loss-prevention policy
dominates a political campaign. We have also seen that a measure of
accountability is achieved by the ability of dissatisfied taxpayers and
businesses to relocate, but the costs of exit are substantial, leaving
taxpayers who must bear the burden of governmental liability with a
far from optimal ability to affect government policy.22 5 Thus, in a re-
gime of no governmental liability, the remaining restraints on gov-
ernment loss-prevention policy are likely to produce less than optimal
results. There is likely to be some marginal benefit from a regime of
governmental liability by enhancing government incentives to invest
in loss prevention, although it is admittedly difficult to estimate its
magnitude.
Accordingly, the case for governmental liability is likely to turn on
the marginal benefits in terms of promoting efficient governmental
investment in loss prevention offered by a regime of governmental
liability. We have seen that many governmental tort immunities op-
erate in areas in which political accountability is likely to be strong-
est-discretionary decisions, the failure to provide adequate police
protection or law enforcement, and the safety of public infrastruc-
ture, for example. In areas where elected officials are most likely to
be held politically accountable for failing to protect the public, the
marginal utility of governmental tort liability is therefore likely to be
small. And from the standpoint of corrective justice, when it is rea-
sonable to expect the public to exact its own form of punishment at
the next election, it is surely problematic to effectively shift liability
onto essentially innocent taxpayers, simultaneously diverting re-
sources from those who may need them far more than the plaintiff
does, and who are essentially blameless for the plaintiffs loss, instead
"4 See, e.g., JERRY L. MASHAW, GREED, CHAOS, AND GOVERNANCE: USING PUBLIC CHOICE TO
IMPROVE PUBLIC LAW 146-47 (1997); DENNIS C. MUELLER, PUBLIC CHOICE II: A REVISED
EDITION OF PUBLIC CHOICE205-06 (1989); NISKANEN, supra note 164, at 135-36.
225 For a critical discussion of the empirical evidence that both casts doubt on the ability of
interjurisdictional competition to produce optimal tax and spending policy while acknowledg-
ing that such competition does constrain the behavior of state and local governments in the
fashion described by the advocates of this view, see William W. Bratton & Joseph A. McCahery,
The New Economics of JurisdictionalCompetition: Devolutionay Federalism in a Second-Best World, 86
GEo. L.J. 201 (1997).
Feb. 2007] A TIEORY OF GOVERNMENTAL DAMAGES LIABILITY 855
of allocating them to a plaintiff who, in most cases, could have

22 6
pur-
chased insurance rather than run a risk of uncompensated loss.
Given the impact of tort liability on republican values as well as
the price paid by essentially innocent third parties who face higher
taxes, reduced government services, or some combination of each as
a consequence of tort liability, statutory tort immunity in areas where
political accountability is likely to operate effectively is well justified.
For similar reasons, caps on recoverable tort damages and the prohi-
bition on punitive damages are equally justifiable. The caps keep
some pressure on elected officials to invest in safety, but mitigate the
hardship on the public at large when scarce government resources
are diverted to the payment of judgments. While the caps mean that
some losses will go uncompensated, without them, other critical
needs of the public may go unaddressed as well. As Part II.C above
explains, governmental tort liability cannot be expected to produce
an efficient result; its virtue lies in its ability to produce some degree
of political pressure on government to invest in loss prevention. Rea-
sonable limitations on governmental damages liability accomplishes
just that result, while mitigating the anomalies that governmental li-
ability can produce.
Discretionary immunity, accompanied by statutory protections on
a public employer's vicarious liability for the torts of its employees, is
also justifiable in light of the problems that inhere in vicarious em-
ployer liability in the public sector. Private firms are liable for the
torts of their employees to ensure that they make all cost-justified in-
vestments in loss prevention and are liable for losses fairly attribut-
able to their own activities, 227 but we have seen that neither efficiency
nor corrective-justice justifications for private-sector liability have sat-
isfactory application in the public sector. Thus, the rule that requires
private-sector employers to essentially act as insurers of the tort liabil-
ity of their employees has no proper application to the public sector.
As long as the government adequately trains and supervises its em-
ployees-a function for which it is likely to be held politically ac-
countable-the justification for requiring the taxpayers to shoulder
the costs of vicarious liability is vanishingly thin.
226 To be sure, not everyone can afford insurance, but there is little reason to believe that the
likelihood that the uninsured will incur uncompensated losses absent governmental tort liability
is greater than the likelihood that the same class of persons will experience reduced benefits
through government programs in a regime of liability. The class suing the government in tort
is largely random; there is little reason to believe that government tort liability is nearly as redis-
tributive as the government programs that would experience reduced funding in a regime of
unlimited liability. The interests of those who cannot afford insurance are unlikely to be ad-
vanced by governmental tort liability.
' See supra text accompanying notes 102, 104.
JOURNAL OF CONSTiTUTIONAL LA W [Vol. 9:3
B. ConstitutionalTorts
1. The Irrelevance of PoliticalAccountability
We have seen that the process of political accountability weakens

the case for governmental liability in common-law tort. The same is
not true for constitutional torts. In that context, the case for substi-
tuting political accountability for a regime of tort liability disappears.
Inherent in the concept of a constitutional right is that its protec-
tion does not depend on the political acceptance of the right at stake.
Thus, political accountability is an unacceptable method for securing
constitutional rights; the Constitution protects even the unpopular or
politically inexpedient. 2 8 Accordingly, discretionary and other cate-
gorical immunities are inappropriate for constitutional torts; a law of
constitutional torts must place pressure on the government to con-
form all of its conduct to the Constitution. That does not imply,
however, that damages are always properly awarded for a constitu-
tional violation. Once one understands that the primary virtue of
damages awards against the government is to create a political incen-
tive to undertake loss prevention, there is ample room for damages-
limiting doctrines that protect the interests of the taxpayers and avoid
unwarranted reallocation of scarce public resources.
2. Qualified Immunity and VicariousEmployer Liability
Part I.B.3 above explains that, given the reality of indemnification,

the stated justification for qualified immunity-avoiding overdeter-
rence of individual public officials-is unpersuasive. That conclusion
does not mean, however, that qualified immunity serves no legitimate
function. When qualified immunity is viewed from the standpoint of
a public employer-the party that bears the economic burden of li-
ability-this doctrine has a compelling justification. Indeed, viewing
qualified immunity from the standpoint of the public employer lends
considerable coherence to this doctrine.
Qualified immunity shields officials from liability unless they vio-
late clearly established law. 2 2 Accordingly, qualified immunity limits
the resources that the government must devote to training and su-
pervision of public officials. Public employers need only undertake
to secure compliance with established legal rules; they need not en-
deavor to train and supervise employees to make more difficult legal
2" Although some advocates of popular constitutionalism take a different view, see, e.g., MARK
TUSHNET, TAKING THE CONSTITUTION AWAY FROM THE COURTS 154-94 (1999), for present pur-
poses it should suffice to observe that their view represents a substantial departure from our
current constitutional regime, see id. at 175-76.
" See supra text accompanying notes 67-68.
Feb.2007] A THEORY OF GOVERNMEN7AL DAMAGES LIABILFTY 857
judgments. Given the expense and uncertain efficacy of training and

supervision directed at unsettled questions of constitutional law, it is
doubtful that the diversion of scarce public resources from other
public purposes toward endeavoring to minimize errors in these un-
settled areas of law-or toward the payment of judgments and other
legal costs when a court concludes that a public official has erred in
such a context-is justifiable.
Consider Wilson v. Layne,2 ° in which the Court, after holding that
the Fourth Amendment forbids law enforcement officers from per-
mitting reporters to accompany them as they execute a search war-
rant for a residence,23' nevertheless granted the officers qualified im-
munity, noting that Fourth Amendment law on this point was
undeveloped when the search at issue occurred and the then-extant
policies of the two law enforcement agencies involved in the case ap-
peared to permit reporters to accompany officers executing a war-
rant.2 3 2 Surely there is no point in holding officers liable for adhering
to office policy. In the absence of any claim against the employees'
supervisors alleging that they had caused the violation by providing
unreasonable training and supervision, liability cannot be expected
to reduce the incidence of constitutional violations. Wilson illustrates
how qualified immunity operates to evaluate the manner in which
public officials have been supervised.
Or consider the decision in Brosseau v. Haugen. In that case, a
police officer shot Haugen, who was wanted on an outstanding felony
warrant, after he had eluded police seeking to execute the warrant,
made his way to his Jeep, and then began to drive off despite the offi-
cer's repeated orders to stop. 234 The Court granted the officer quali-
fied immunity on the ground that no clear rule had emerged about
whether the Constitution permits an officer "to shoot a disturbed
felon, set on avoiding capture through vehicular flight, when persons
in the immediate area are at risk from that flight." 5 Justice Stevens,
in dissent, nevertheless made a compelling case that the officer's
conduct was unreasonable; Haugen had not committed a violent
crime or threatened the officer, and there was no evidence that he
intended to drive off recklessly or otherwise endanger bystanders.236
Justice Stevens acknowledged, however, that there was some risk that
Brosseau might have injured someone as he drove off, and therefore
that there was "uncertainty about how a reasonable officer making
" 526 U.S. 603 (1999).

211 Id. at 611-13.
22 Id. at 615-18.
213543 U.S. 194 (2004) (per curiam).
21 Id. at 194-97.
Id.at 200.
2 I5
See id. at 202-07 (Stevens, J., dissenting).
JOURNAL OF CONSTIUTIONAL LAW [Vol. 9:3
the split-second decision to use deadly force would have assessed the
foreseeability of a serious accident .... But in this context, it is
doubtful that any greater investment in training and supervision of
officers could have prevented a constitutional violation. The appli-
cable constitutional rule is that a fleeing felon can be shot only when
there is "probable cause to believe that the suspect poses a threat of
serious physical harm, either to the officer or to others... ,,238It is
surely reasonable to expect public employers to train law enforce-
ment officers to know this rule, but it is unclear that additional train-
ing and supervision will enhance the likelihood that the officer will
act correctly when making a split-second decision about whether a
fleeing felon is likely to endanger someone else. A significant error
rate is inherent in split-second police judgments, and when a law en-
forcement agency has invested all resources that are reasonably war-
ranted in training and supervision with respect to constitutional re-
quirements, the case for awarding damages does not persuade; the
damages award will not likely reduce the rate of constitutional viola-
tions, but it would divert public resources that could be invested far
more efficaciously in other types of public goods, including loss pre-
vention.2 9
Qualified immunity therefore properly shields public officials
from liability in cases where the public employer has made reason-
able investments in securing compliance with the Constitution.
Qualified immunity places the burden on the employer-indemnitor
to monitor its employees with respect to clearly established law, the
context in which monitoring is most likely to be cost-effective. While
phrased as a protection for public employees, qualified immunity, as
it operates, judges the reasonableness of the employer's conduct by
giving it an incentive to reduce constitutional injuries where the law
is settled and the costs of monitoring are therefore reasonable. Mu-
nicipal liability is also appropriate when based on a culpable munici-
pal policy; once again, liability is properly premised on the govern-
ment's own failure to take sufficient measures to ensure compliance
with the Constitution.4 There is, however, little justification for in-
"' Id. at 206. Believing such police decisions to present fact-specific issues, Justice Stevens
maintained that the case should be tried. See id. at 205-08.
Id. at 197-98 (quoting Tennessee v. Garner, 471 U.S. 1, 11 (1985)).
... Even before Brosseau, the Court had held that qualified immunity protects officials even
when they are applying relatively settled legal standards but nevertheless are required to make
difficult judgments as applied to the particular facts confronting the official. See Saucier v. Katz,
533 U.S. 194, 197 (2001); Anderson v. Creighton, 483 U.S. 635, 641 (1987).
" See supra text accompanying note 81. I have no quarrel with Justice Breyer's observation
that there is little point to imposing different liability rules for municipalities and municipal
employees inasmuch as the former usually indemnify the latter. See Bd.of County Comm'rs v.
Brown, 520 U.S. 397, 436 (1997) (Breyer, J., dissenting). But because virtually all constitutional
tort liability in reality resides with employers, qualified immunity serves the same function as the
Feb.2007] A THEORY OFGOVERNMENTAL DAMAGES LIABILrTY 859
sisting on a greater allocation of public resources to loss prevention

when the ability to reduce constitutional violations with respect to
unsettled law or unclear legal obligations is itself open to great
doubt. 24' In the face of unsettled legal obligations, the most likely ef-
fect of imposing liability is to require a public employer to provide
what amounts to constitutional-tort insurance. We have seen, how-
ever, that a regime of vicarious employer liability has little justifica-
tion in the public sector, where tort law cannot hope to produce the
efficient level of investment
responeat . 242 in loss prevention that is the objective of
respondeat superior. Thus, constitutional tort liability produces po-
litical pressure on government to conform its conduct to the Consti-
tution, while qualified immunity and limited employer liability bar
requirement of a culpable policy by limiting liability to cases in which the employer failed to
take reasonable measures to avoid constitutional violations. This approach suggests that state
and federal governments should face policy liability for constitutional torts-a result neverthe-
less forbidden by sovereign immunity-and that municipal liability might properly be based on
negligence rather than deliberate indifference. But it is unclear whether these limitations on
governmental liability have much practical significance inasmuch as a public employee can
generally be named as a defendant in a constitutional tort action, even when the employer is
immune or otherwise not liable, and the defense of qualified immunity limits liability to cases
involving fairly culpable employers. For a similar argument that qualified immunity makes con-
stitutional tort liability properly turn on culpability, see Jeffries, supra note 55, at 54-59. In-
deed, one could argue that policy liability for local governments should be abolished; given the
unavailability of qualified immunity, even a municipal policy that represents a reasonable effort
to construe uncertain constitutional obligations, such as the policies considered in Wilson v.
Layne, could result in damages liability.
241 Some claim that qualified immunity hinders the development
of constitutional law be-
cause plaintiffs may be denied recovery in cases such as Wilson in which they press a novel con-
stitutional claim. See, e.g., Brown, supra note 193, at 1101-10. But the empirical case that quali-
fied immunity has stunted the development of constitutional law has yet to be made. Despite
qualified immunity, new constitutional law grows from cases seeking injunctive relief and attack-
ing municipal policies, as well as in criminal litigation and in cases like Wilson itself. Moreover,
the development of constitutional law is facilitated by the Court's practice of reaching the mer-
its before considering whether a damages award is defeated by qualified immunity. See Wilson
v. Layne, 526 U.S. 603, 609 (1999); County of Sacramento v. Lewis, 523 U.S. 833, 841 n.5
(1998). See generally John M.M. Greabe, Mirabile Dictum!: The Case for "Unnecessary" Constitu-
tionalRulings in Civil Rights Damages Actions, 74 NOTRE DAME L. REv. 403 (1999) (defending the
practice of reaching the merits even in actions barred by qualified immunity). I am unaware of
any empirical evidence that the doctrine of qualified immunity has operated to inhibit the de-
velopment of constitutional law.
242 Accordingly, although I disagree with much of the Court's
reasoning, in my view the
Court correctly denied private firms managing public prisons qualified immunity in Richardson
v. McKnight, 521 U.S. 399 (1997). A private firm is subject to market discipline and therefore
should internalize the costs of its employees' misconduct so that the firm's services are priced to
enable it to engage in cost-justified loss prevention. A publicly run prison, in contrast, is subject
to political and not market discipline, and therefore should not face constitutional tort liability
when it undertakes reasonable efforts to achieve compliance with constitutional norms.
Richardson is a close case, however, because public contractors have a degree of political ac-
countability; they must fear losing their contracts if their performances produce adverse politi-
cal fallout.
awards of damages in contexts where liability is unlikely to reduce the

incidence of constitutional violations.
3. Measure ofDamages
These same considerations support the refusal to base constitu-
tional tort damages on the presumed or inherent value of constitu-
tional rights. 243 Aside from the difficulty of identifying a nonarbitrary
basis to measure the abstract value of a constitutional right, imposing
this type of liability would be unacceptably likely to overdeter; pre-
sumed damages awards inevitably place pressure on the government
to engage in even non-cost-justified loss-prevention measures. It
would be far better to achieve deterrence through a regime of puni-
tive damages collectable from individual wrongdoers than through
awarding a form of presumed damages that is ultimately paid by the
taxpaying public. Indeed, it was my experience in municipal gov-
ernment that nothing terrified my clients more than the possibility of
an award of punitive damages, for which they could not be indemni-
fied. When I wanted to impress upon an official, even at the highest
levels of municipal government, the magnitude of legal risk inhering
in a particular course of action, nothing worked better than a discus-
sion of punitive damages. Given the costs that governmental liability
imposes on essentially innocent third parties, punitive damages paid
by the actual wrongdoer are far preferable to a regime of presumed
compensatory damages that would inevitably be punitive in effect.24
Indeed, the Supreme Court has made just this point as it rejected the
availability of punitive damages against municipalities under Section
1983.246 Moreover, punitive damages represent the ideal solution for
officials who are willing to countenance or even encourage constitu-
tional violations because of the political benefits they yield. Punitive
damages are thought warranted in the private sector when compensa-
tory awards are likely to give defendants insufficient incentives to
avoid tortious behavior;246 in the public sector, punitive damages are
warranted when compensatory awards may provide officials with in-
23 See supra text accompanying note 65.

2 Although the Supreme Court has yet to decide the question, the emerging consensus in
the lower courts is that punitive damages may be awarded for a constitutional tort even in the
absence of compensatory damages. See, e.g., Kelly Koenig Levi, Allowing a Title VI PunitiveDam-
ages Award Without an Accompanying Compensatory or Nominal Award: Further Unifying Federal Civil
Rights Law, 89 KY. L.J. 581, 595 (2000-01).
2'45See City of Newport v. Fact Concerts, Inc., 453 U.S. 247, 269-70 (1981).
246 See, e.g., Robert D. Cooter, PunitiveDamagesfor Deterrence: When and How Much?, 40 ALA. L.
REV. 1143, 1144-48 (1989); Dorsey D. Ellis, Jr., Fairness and Efficiency in the Law of Punitive Dam-
ages, 56 S. CAL. L. REV. 1, 77-78 (1982); Thomas C. Galligan, Jr., Augmented Awards: The Efficient
Evolution of PunitiveDamages, 51 LA. L. REV. 3, 11-12 (1990); Keith N. Hylton, Punitive Damages
and the Economic Theory of Penalties,87 GEO. L.J. 421,423 (1998).
Feb. 2007] A THEORY OF GO ERNMErIVAL DAMAGES LIABILITY 861
sufficient political incentive to comply with the Constitution. Puni-

tive damages are appropriate when compensatory damages will un-
derdeter, and therefore punitive damages are properly awarded in
cases in which compensatory damages do not overcome a political in-
centive to violate the Constitution.
4. Statutory Limitations on Damages

The Bivens line of cases suggests that damages liability for consti-
tutional torts is constitutionally compelled, at least absent a satisfac-
tory alternative remedy or the presence of special factors, since a con-
stitutional right without a correlative remedy is no right at all.248 The
Court has provided little guidance, however, as to what constitutes an
adequate alternative remedy.249 It follows from the view advanced
here that an adequate remedy should ordinarily involve a financial
consequence for a constitutional violation to ensure that it imposes a
247 The clear majority of states do not authorize indemnification of public employees for
egregious misconduct or punitive damages. See statutes cited supra note 51. Moreover, al-
though most courts that have considered this question have ruled that federal law permits in-
demnification of public employees for punitive damages, see Schwartz, supra note 89, at 1219-
23, it may well be that, on the view advanced here that the purpose of a punitive-damages award
is to negate political incentives to commit (or at least tolerate) constitutional violations without
imposing costs on essentially innocent third parties-the taxpayers and those dependent on
public services-federal law preempts state laws permitting indemnification for constitutional-
tort punitive-damage awards. If federal law is properly understood to have as its objective the
creation of an individual-employee incentive to avoid constitutional violations through placing
their personal assets at risk, then state indemnity statutes would be inconsistent with this federal
objective and would therefore be preempted. See Crosby v. Nat'l Foreign Trade Council, 530
U.S. 363, 373 (2000); Livadas v. Bradshaw, 512 U.S. 107, 120 (1994); Hines v. Davidowitz, 312
U.S. 52, 67 (1941).
242 See Correctional Servs. Corp. v. Malesko,
534 U.S. 61, 72-74 (2001); Mitchell v. Forsythe,
472 U.S. 511, 522-23 (1985); Carlson v. Green, 446 U.S. 14, 18-25 (1980); Bivens v. Six Un-
known Named Agents, 403 U.S. 388, 395-97 (1971).
2" See generally, e.g., Susan Bandes, Reinventing Bivens: The Self-Executing Constitution, 68 S.
CAL. L. REV. 289, 329-36 (1995) (discussing the nature of the remedy for constitutional viola-
tions by federal officials and concluding that in some cases courts may choose among reme-
dies); George D. Brown, Letting Statutory Tails Wag ConstitutionalDogs: Have the Bivens Dissenters
Prevailed?, 64 IND. L.J. 263, 269-74 (1989) (noting that the Court has concluded that other
remedies for federal constitutional violations given by Congress can preclude any right to dam-
ages); Dan T. Coenen, A Constitution of Collaboration: ProtectingFundamental Values with Second-
Look Rules of Interbranch Dialogue, 42 WM. & MARY L. REV. 1575, 1747-49 (2001) (discussing the
necessity to think through the structural rules that exist in ourjurisprudence as the pressures to
apply them increase in complex litigation); Betsy J. Grey, Preemption of Bivens Claims: How
Clearly Must Congress Speak?, 70 WASH. U. L.Q. 1087, 1089-97 (1992) (reviewing the evolution of
the Court's acceptance of congressional remedies as an alternative to damage remedies for con-
stitutional violations by federal officers); Gene R. Nichol, Bivens, Chilicky, and Constitutional
Damages Claims, 75 VA. L. REV. 1117, 1138-45 (1989) (discussing alternative remedies to dam-
ages); Note, Bivens Doctrine in Flux: Statutory Preclusion of a Constitutional Cause of Action, 101
HARV. L. REV. 1251, 1258-60 (1988) (advocating a constitutional requirement of effective reme-
dies to vindicate rights).
political price. On this view, the decision in Schweiker v. Chilicky, at

first blush, seems problematic. In that case, the Court rejected a
Bivens claim for consequential damages caused by wrongful denial of
disability benefits on the ground that Congress had authorized no
more than an award of retroactive benefits in the disability-benefits
51
provisions of the Social Security Act.2 As a result, there was no
budgetary impact associated with a constitutional violation, which is
seemingly at odds with the view advanced here. But as I have en-
deavored to demonstrate, public-sector liability rules should be sensi-
tive to the impact of liability on third parties and the provision of
government services. These considerations offer a defense for
Chilicky.
The constitutional violation alleged in Chilicky was that administra-
tors had deprived the plaintiffs of property without due process of law
by manipulating the disability review process to wrongfully terminate
benefits in order to reduce the cost of the program. The statutory
remedy of retroactive benefits, however, prevented administrators
esrc.253
from achieving the budgetary reductions they desired. For that rea-
son, the remedy deprived them of the political benefit they sought in
reducing expenditures on an evidently disfavored constituency. A
statutory remedy that deprives administrators of the political benefits
they seek from violating the law may well be sufficient to sustain the
adequacy of that remedy-especially when coupled with the availabil-
ity of punitive damages against individual officials responsible for
weakening the integrity of the program. Aside from that, the alleged
misconduct involved undermining a congressional intent to provide
the benefits at issue; the violation was therefore subject to what was
likely to be effective political scrutiny. Indeed, by the time the case
reached the Court, Congress had already overhauled the program to
eliminate perceived abuses.' 4 Moreover, the Court was undoubtedly
right to concern itself with the potential for budgetary disruption
caused by the• •€, of• unlimited
threat
• 155 liability for officials who are likely
to require indemnification. If unlimited damages were available for
a wrongful denial of disability benefits, the federal disability program
could become so expensive that political support for it would ebb.
This would hardly strike a blow for the disabled or others dependent
on government aid.
' 487 U.S. 412 (1988).

...Id. at 424-29.
252id. at 415-20.
25 See id. at 417-18.
'" Id. at 415-17.
...Id. at 425.
For similar reasons, statutory damages caps like those many states
have enacted for common-law torts are defensible. We have seen that
reasonable damages caps preserve political pressure on government
to conform its conduct to the law but mitigate the anomalies associ-
ated with governmental damages liability. Bush v. Lucas56 provides an
example: the Court rejected a Bivens action in light of the availability
of statutory civil-service remedies that granted the successful plaintiffs
retroactive seniority and backpay.5 7 Since backpay and retroactive
seniority mean that the government, in effect, must pay an employee
for not working, there is a significant political cost to these remedies
that should sustain their adequacy. At a minimum, an inquiry into
the political efficacy of a statutory remedy will usefully guide an as-
sessment of its adequacy.
Thus, a focus on the political costs associated with statutory limita-
tions on constitutional tort damages provides a useful vehicle for
evaluating their adequacy in light of Bivens's objective of ensuring
that constitutional rights have correlative remedies. A regime of lim-
ited liability that nevertheless imposes a sufficient political price to
minimize the likelihood of constitutional violations should be sus-
tained.
C. Takings
1. The Role of Compensation in Limiting Overuse of Eminent Domain
We have seen that the function of the constitutional requirement

of compensation is to impose political restraint on the power of emi-
nent domain. The upshot of the compensation requirement is that
government will not exercise its power of eminent domain except
when the political benefits of condemnation exceed the political cost
of compensation. Consideration of the political operation of a re-
gime in which a governmental taking did not require compensation
makes the point.
Absent a constitutional requirement of compensation, property
owners could protect their investments by securing takings insurance,
and the rising cost of takings insurance that would result from over-
use of condemnation could create a political restraint on elected offi-
cials.2 58 But it is unclear how effectively insurance rates would check
2 462 U.S. 367 (1983).

id. at 388-89.
The availability of insurance and the political impact of high insurance rates caused by
promiscuous use of eminent domain has caused some to argue that the compensation require-
ment is unjustified. See Steven P. Calandrillo, Eminent Domain Economics: Should 'Just Compensa-
tion" Be Abolished, and Would "Takings Insurance" Work Instead?, 64 OHIO ST. L.J. 451 (2003);
Louis Kaplow, An Economic Analysis of Legal Transitions,99 HARV. L. REv. 509, 602-06 (1986).
governmental overuse of condemnation. After all, the cost of insur-

ance is hardly the only issue on which the public judges politicians,
and we have seen that political accountability operates imperfectly.
The requirement of compensation, in contrast, imposes far more di-
rect political discipline on the use of eminent domain. The govern-
ment must budget for compensation and incur the political opportu-
nity costs associated with its payment.259 Indeed, the compensation
requirement imposes a far more direct political restraint than does
ordinary tort liability. The incentive to invest in loss prevention cre-
ated by common-law or constitutional tort liability is muted by the
difficulty of predicting future liabilities and the extent to which they
will be reduced by investment in loss-prevention measures. 2w The
compensation requirement affects government budgets in a far more
immediate and predictable fashion-compensation is required for
every taking, compensation must generally be paid at the time of the
taking, and its cost is readily determinable through appraisal. 26' Thus
the political impact of compensation is unusually direct, immediate,
and predictable.
2. The Public-Use Inquiry

The preceding discussion suggests that the Supreme Court cor-
rectly defers to the judgment of elected officials when deciding
.2 Although they have not linked the observation to a theory of political behavior like the
one advanced above, commentators have observed that the budgetary impact of compensation
will restrain the use of eminent domain. See, e.g., FISCHEL, supra note 200, at 73-75; Thomas W.
Merrill, The Economics of Public Use, 72 CORNELL L. REv. 61, 77-81 (1986). Professor Fischel,
however, has argued that this discipline is undermined when another unit of government in
significant part finances eminent domain, such as when a municipal government uses a federal
grant to cover the cost of condemnation. William A. Fischel, The PoliticalEconomy of Public Use in
Poletown: How Federal Grants EncourageExcessive Use of Eminent Domain, 2004 MICH. ST. L. REV.
929, 943-46. Professor Fischel, however, overlooks political opportunity costs, not only for local
officials when they use their political capital to lobby for federal funding of eminent-domain
costs as opposed to some other project, but also for members of Congress when they allocate
scarce public resources.
2o See supra note 186.
It has long been settled that the Constitution forbids a taking when compensation is not
paid in advance, unless the property owner has available a means of obtaining compensation
with reasonable certainty and without unreasonable delay. See Bragg v. Weaver, 251 U.S. 57, 62
(1919); 3JULIUS L. SACKMAN, NICHOLS ON EMINENT DOMAIN § 8.05(2) (2006). And aside from
this constitutional requirement, under federal law, title does not pass until compensation has
been paid. See Kirby Forest Indus., Inc. v. United States, 467 U.S. 1, 11-13 (1984). Also, for
federal condemnations or state or local condemnations that receive federal financial assistance,
the condemner cannot take possession unless it has paid the agreed purchase price or depos-
ited with the court an amount not less than the compensation specified in an approved ap-
praisal or the amount awarded in a condemnation proceeding. See 42 U.S.C. § 4651 (4) (2000).
Under the Takings Clause itself, if the government takes property before compensation is paid,
it is liable for prejudgment interest on the requisite compensation. See Kirby Forest Indus., 467
U.S. at 10-11.
whether a taking is for a "public use" within the meaning of the Tak-
ings Clause-even when property is condemned for the purpose of
transferring it to a private party to promote economic development
as in Kelo v. City of New London.
The Takings Clause's public-use requirement is, at best, ambigu-
ous. Almost no one believes that a taking for "public use" occurs only
when the public will physically traverse the condemned property-
such reasoning would prohibit takings in order to build prisons, mili-
tary bases, or other facilities not open to the public. Instead, even the
most vigorous advocates of restricting eminent domain resist this view
and acknowledge that the public can "use" a facility within the mean-
ing of the Takings Clause, even when it is not generally open to the
public, as long as the public receives some sort of benefit from gov-
ernmental ownership of the property. 263 Once this type of indirect
and metaphorical public use is accepted as consistent with the consti-
tutional text, however, it becomes difficult to draw lines. In some
sense the public "uses" the benefits of redevelopment that lowers
crime and generates new tax revenues just as it "uses" prisons and
military bases.2 6 Nor can the question of public use plausibly turn on
whether the government holds title to the property after its condem-
nation. The public makes the same metaphorical "use" of toll roads
or utilities that have been acquired by condemnation whether they
are publicly or privately owned. 265 For this reason, the opponents of
condemnation for the benefit of a private party are forced to advo-
cate all sorts of complicated tests to decide whether a particular con-
demnation for the purpose of eventual transfer to a private party is
for a public use in the constitutional sense. 266 Litigation governed by
2 545 U.S. 469 (2005); see also Haw. Hous. Auth. v. Midkiff, 467 U.S. 229, 245 (1984) (ap-
proving the taking of land from an oligopolist for redistribution); Berman v. Parker, 348 U.S.
26, 35-36 (1954) (approving a taking for the purpose of slum clearance and private redevelop-
ment).
20 See, e.g., RICHARD A. EPSTEIN, TAKINGS: PRIVATE PROPERTY AND THE POWER OF EMINENT
DOMAIN 167-68 (1985).
For an argument about the difficulty of drawing principled distinctions in this area, see
Thomas Ross, TransferringLand to Private Entities by the Power of Eminent Domain, 51 GEO. WASH.
L. REV. 355, 369-74 (1983).
2'6 The use of eminent domain to acquire property for use by a privately owned utility is well
settled. See, e.g., Hendersonville Light & Power Co. v. Blue Ridge Interurban Ry. Co., 243 U.S.
563, 568-70 (1917) (approving taking of property for construction of a dam operated by a util-
ity company); Mt. Vernon-Woodberry Cotton Duck Co. v. Ala. Interstate Power Co., 240 U.S.
30, 32-33 (1916) (sanctioning a public taking of property to enable a power company to build a
dam).
' See, e.g., Kelo, 545 U.S. at 497-502 (O'Connor, J., dissenting) (permitting condemnation
for the benefit of a private party when the property will be available for the public's use or when
the existing property inflicts an affirmative harm defined broadly to include blight or oligop-
oly); County of Wayne v. Hathcock, 684 N.W.2d 765, 781-83 (Mich. 2004) (en banc) (holding
that a condemnation is for public use when it is a public necessity, when a private party receiv-
ing property is still accountable to the public, or when the land is selected on the basis of a pub-
such standards would necessarily be complex, expensive, and prone

to error.
Nor does inquiry into the original meaning of the Takings Clause
resolve the confusion. A case can be made that the Clause was in-
tended to forbid condemnation for use by a private party,267 but there
is also a strong case to be made that such condemnations are consis-
tent with the original meaning of eminent domain, based on the his-
torical evidence that condemnation by or for the benefit of private
parties was considered permissible when those parties provided a util-
ity-type service that was considered to be of benefit to the public at
large.2 68 Inquiry into original meaning is further complicated by the
fact that the Takings Clause does not literally prohibit takings
thought not to be for "public use"; instead it treats the existence of
"public use" as a trigger for compensation. Thus, it is plausible to
read the Takings Clause as doing no more than identifying a type of
taking for which compensation is required (a taking for "public
use"), while other kinds of takings of property by the government,
such as taking by tort, taxation, or police-power regulation, were not
forbidden but merely excluded from the compensation require-
ment.269 Indeed, unfamiliar though this argument is to many, it is
largely uncontroversial. No one thinks that the Takings Clause ap-
plies to all takings of private property by the government; no one ar-
gues, for example, that the government must make compensation
when it takes property through criminal fines or forfeiture. Thus, it
is unclear that the Takings Clause should be read to forbid takings
that are not for a "public use."
Given the many difficulties with assessing the meaning and import
of the public-use requirement, the opponents of takings for purposes
lic concern); Lee Anne Fennell, Taking Eminent Domain Apart, 2004 MICH. ST. L. REV. 957, 987-
92 (advocating an inquiry into whether condemnation places an undue burden on the interests
of property owners); Garnett, supra note 142, at 963-82 (advocating inquiry into nexus and pro-
portionality between condemnation and purported use to determine if use of eminent domain
is appropriate).
26 See Kelo, 545 U.S. at 506-14 (Thomas, J., dissenting); Eric R. Claeys, Public-UseLimitations
and NaturalProperty Rights, 2004 MICH. ST. L. REV. 877, 892-900.
2" See, e.g., Lawrence Berger, The Public Use Requirement in Eminent Domain, 57 OR. L. REV. 203,
204-12 (1978); Charles E. Cohen, Eminent Domain After Kelo v. City of New London: An Argu-
ment for Banning Economic Development Takings, 29 HARV. J.L. & PUB. POLY 491, 500-10 (2006);
Buckner F. Melton, Jr., Eminent Domain, "Public Use," and the Conundrum of Original Intent, 36
NAT. RESoURCESJ. 59, 68-80 (1996).
' See Matthew P. Harrington, "PublicUse" and the Original Understandingof the So-Called "Tak-
ings" Clause, 53 HASTINGS L.J. 1245, 1278-301 (2002);Jed Rubenfeld, Usings, 102 YALE L.J. 1077,
1119-24 (1993).
20 The Supreme Court, for example, has explained that the government is not "required to
compensate an owner for property which it has already lawfully acquired under the exercise of
governmental authority other than the power of eminent domain." Bennis v. Michigan, 516
U.S. 442, 452 (1996).
of private redevelopment generally make their case by arguing that

such takings are unacceptably likely to amount to no more than a
transfer of wealth to a politically favored private party. 7' Of course,
the Takings Clause does not prohibit a taking merely because private
parties will derive benefit from the taking-elected officials presuma-
bly undertake eminent domain nearly always because it will benefit at
least some of their constituents. The constitutional text requires that
the taking be for "public use," not that it fail to benefit private par-
ties. And once a metaphorical "public use" of property not physically
open to the public is accepted as triggering the compensation re-
quirement, the theory advanced here suggests that takings will nearly
always satisfy the metaphorical test. It is unlikely that elected officials
would choose to deploy scarce government resources in a fashion
that yields no benefit to anyone but the developer. The theory of po-
litical behavior advanced above suggests that the allocation of scarce
public resources will be aimed at assembling winning electoral coali-
tions-not mere transfers of public wealth to favored parties. If an
elected official allocates resources for no reason other than to benefit
a favored private interest, the likelihood of retribution in the next
election should be, at least for most elected officials, unacceptably
high, and the political opportunity costs of utilizing the money spent
on condemnation in such a politically disadvantageous manner
would be enormous.
Accordingly, the theory advanced here argues for deference to po-
litical judgments about condemnation. The Supreme Court ordinar-
ily defers to legislative decisions in the area of social and economic
policy because the process of political accountability is considered the
most reliable method for rectifying improvident decision-making. 72
There is little reason to doubt the applicability of this insight when it
comes to eminent domain. Indeed, as we have seen, the compensa-
tion requirement imposes political discipline well in excess of that
usually operating in the political arena. If, as I have argued, the pur-
' See, e.g., EPSTEIN, supra note 263, at 169-70; Cohen, supranote 268, at 543-50; StephenJ.
Jones, Trumping Eminent Domain Law: An Argument for Strict Scrutiny Analysis Under the Public Use
Requirement of the Fifth Amendment, 50 SYRACUSE L. REv. 285, 287-88 (2000); Gideon Kanner,
Kelo v. New London: Bad Law, Bad Policy, and Bad Judgment, 38 URB. LAW. 201, 206-09 (2006);
Daniel B. Kelly, The "PublicUse" Requirement in Eminent Domain Law: A Rationale Based on Secret
Purchases and Private Influence, 92 CORNELL L. REV. 1, 34-41 (2006); Donald J. Kochan, "Public
Use" and the IndependentJudiciary: Condemnation in an Interest-Group Perspective, 3 TEX. REV. L. &
POL. 49, 69-72 (1998); Ilya Somin, Overcoming Poletown: County of Wayne v. Hathcock, Eco-
nomic Development Takings, and the Futureof Public Use, 2004 MICH. ST. L. REv. 1005, 1006-09.
' As the Court has put it, "[tihe Constitution presumes that, absent some reason to infer
antipathy, even improvident decisions will eventually be rectified by the democratic process and
that judicial intervention is generally unwarranted no matter how unwisely we may think a po-
litical branch has acted." FCC v. Beach Commc'ns, Inc., 508 U.S. 307, 314 (1993) (quoting
Vance v. Bradley, 440 U.S. 93, 97 (1979)).
JOURNAL OFCONSTTITTIONAL LA W [Vol. 9:3
pose of the compensation requirement is to impose political disci-

pline on the exercise of the power of eminent domain, then that
purpose is satisfied regardless of whether the taking is for purposes of
private redevelopment-the political discipline itself is the best guar-
antee that the taking will serve the public at large.
Moreover, the political discipline imposed by the compensation
requirement is far more likely to result in reliable determinations
about the public benefits expected from a redevelopment project
than a regime of intensive judicial review. After all, we can expect
elected officials to be more likely to have relevant expertise on what is
essentially a policy (and not a legal) question about what benefits will
flow from a proposed redevelopment project-and far more incen-
tive to identify only those projects likely to yield sufficiently wide-
spread benefits to have political utility in upcoming elections.
3. The StandardforJust Compensation

We have seen that the constitutional requirement of compensa-
tion cannot be expected to produce only costjustified takings of pri-
vate property because the political costs of paying the requisite com-
pensation cannot readily be monetized. Thus, those who advocate
any number of nicely calibrated adjustments in the level of compen-
sation in order to achieve greater efficiency in condemnation are en-
gaged in a hopeless endeavor. 73 One can debate what level of com-
pensation is 'just" from the standpoint of the property owner, but
because the political costs of compensation cannot readily be
monetized, there is no ready way to use the standard of compensation
to achieve an "efficient" level of takings. Similarly, in attempting to
assess the 'just" level of compensation, one cannot focus exclusively
on the interests of the property owner. As we have seen, the interests
of a variety of third parties are at stake when the government is re-
quired to make compensation. While the theory of governmental
damages liability advanced here does not argue for any particular
standard of compensation, it does suggest that a 'just" outcome can-
not be reached if the interests of the taxpayers and those who are de-
pendent27 4
on government-funded services are not also taken into ac-
count.
s The ordinary measure of just compensation is the fair market value of the parcel at the
time of the taking. See, e.g., United States v. 50 Acres of Land, 469 U.S. 24, 29 (1984); Kirby
Forest Indus., Inc. v. United States, 467 U.S. 1, 10 (1984). For a recent survey of the academic
debate over the appropriate standard forjust compensation, see Serkin, supra note 142, at 687-
703.
274 A comprehensive discussion of the appropriate benchmark for just compensation is well
beyond the scope of this Article, but it is worth noting the claim that the current standard of
fair market value provides undercompensation because owners frequently place subjective value
In one respect, however, the theory advanced here has significant

implications for the measure of compensation. As we have seen, the
function of the Takings Clause is to place a political restraint on the
use of eminent domain. Accordingly, the measure of 'Just compensa-
tion" when property is condemned for the purposes of transfer to a
private party should differ from the compensation owed for property
215
that the government acquires for its own use.
The condemning authority, when it assembles a large parcel for
conveyance to a developer, may receive a price in excess of its costs of
acquiring individual parcels; a large parcel invites a variety of inten-
sive commercial or industrial uses that may possess far greater value
than the price paid for the individual parcels from which it was as-
sembled. Indeed, the need to assemble large parcels coupled with
the holdout problems caused by the owners of individual parcels is
one of the primary justifications for eminent domain. In such
cases, by offsetting its acquisition costs as it resells the property, the
condemning authority mitigates the budgetary impact of condemna-
tion; it could even turn a profit. In this fashion, the political check
on overuse of eminent domain created by the compensation re-
quirement is undermined. But when one understands the just-
compensation requirement in terms of the political cost that the Tak-
ings Clause requires to be exacted in connection with condemnation,
it follows that the requisite 'just compensation" should include the
premium that the government has or may obtain through a redevel-
opment project. This scheme would properly preserve the political
restraint imposed by the compensation requirement. And if, as I
have argued, the compensation requirement is properly understood
as a political restraint on takings, then it is essential that government
not dilute this political restraint by turning eminent domain into
on retaining their property not valued by the market. See, e.g., Michael DeBow, Unjust Compensa-
tion: The ContinuingNeed for Reform, 46 S.C. L. REv. 579, 591 (1995); Fennell, supra note 266, at
962-67. The theory advanced here, however, urges consideration of the interests of third par-
ties that are affected by governmental damages liability, and in that connection it is equally
worth noting that a measure of compensation that includes a subjective component would en-
courage strategic behavior and produce substantial transaction costs as subjective valuation is
litigated. Political accountability should also reduce undercompensation because of the politi-
cal costs associated with what could be perceived as governmental overreaching. Moreover, if
undercompensation were a systemic problem, we should then expect the private market to offer
takings insurance. The fact that no such market has appeared is at least some indication that
complaints about undercompensation are exaggerated. For an empirical case that undercom-
pensation concerns are overstated, see Nicole Stelle Garnett, The Neglected Political Economy of
Eminent Domain, 105 MICH. L. REV. 101 (2006).
171 In Kelo, the Court noted but reserved decision on the question
of what the appropriate
measure ofjust compensation should be in cases involving condemnation for redevelopment by
a private party. See Kelo v. City of New London, 545 U.S. 469, 489 n.21 (2005).
7 SeeJames E. Krier & Christopher Serkin, Public Ruses, 2004 MICH.
ST. L. REv. 859, 872-73.
17 See, e.g., POSNER, supra note 94, at 56-57; Merrill, supra note 259, at 74-77.
what could amount to a revenue center.
CONCLUSION
Government responds to political, not market, signals. It should

therefore come as no surprise that economic theory offers limited in-
sight into governmental tort liability. And because government ex-
ternalizes its costs, corrective justice also offers little insight into gov-
ernmental tort liability. Government operates in a political context,
and accordingly governmental tort liability is best understood in po-
litical terms. The Supreme Court endeavored to offer such a theory
in Richardsonv. McKnight, but its account was seriously deficient.
My ambition is to argue the case for and against governmental
tort liability with equal vigor. Tort liability imposes a serious cost on
elected officials intent on deploying public resources to maximum
political advantage. The fashionable academic skepticism about the
efficacy of governmental tort liability is therefore quite misguided.
But governmental tort liability also imposes a cost on innocent par-
ties-not just the taxpayers, but also on the most vulnerable among
us who are generally in greatest need of government assistance and
most likely to lose out when public resources are diverted to the de-
fense of litigation and the payment of judgments. An attractive ac-
count of governmental tort liability must pay close attention to both
of these costs. Tort liability properly restrains governmental miscon-
duct; but too much of a good thing usually becomes a bad thing, and
tort liability is no exception. In an attractive and just regime of gov-
ernmental tort liability, well-tailored immunity rules are no less essen-
tial than a measure of liability itself.

A Theory of Governmental Damages Liability - Torts Constitutional

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Theory of Governmental Damages Liability - Torts Constitutional

Uploaded by

Copyright:

Available Formats

A THEORY OF GOVERNMENTAL DAMAGES LIABILITY:

TORTS, CONSTITUTIONAL TORTS, AND TAKINGS

INTRO DU CTIO N .................................................................................... 798

L iab ility ............................................................... 856

Governmental damages liability in tort represents a special case, to

Associate Professor of Law, Chapman University School of Law. I received assistance

supplies little support for public-sector tort liability. 4 Indeed, there is

4 See infra text accompanying notes 122-24.

to the defense of litigation and the payment of judgments has impor-

I. THE CONTOURS OF GOVERNMENTAL TORT LIABILITY

7 545 U.S. 469 (2005).

ability.1 3 The view that sovereign immunity has no constitutional

the government to pay a judgment absent legislative authorization for

16 183 U.S. 62 (1901).

2. State andLocal Liability

2' 28 U.S.C. § 2674 (2000).

Many state immunity statutes confer immunities on governmental de-

employees,3 1 and forty states confer immunity from punitive dam-

ernments or their employees include immunity for issuance, denial,

administrative proceedings; 31 the plan or design for public improve-

other corrections or penal facilities; 42 the probation, parole, release,

property-and otherwise grant immunity. 8 In addition, forty-two

State statutes usually grant public employees additional protec-

3. Asymmetry with Private Tort Liability

without fear of liability. It should therefore be no surprise that the

power to engage in tortious conduct without paying the injured party

Liability for constitutional torts turns on whether the defendant is

Sovereign immunity bars an action seeking to recover damages

on the presumed or intrinsic value of constitutional rights; they can

2. State and Local Liability

U.S. 247, 253-58 (1978).

however, construes the Eleventh Amendment to permit Congress to

liberate indifference to constitutional violations by municipal em-

3. Asymmetry with Private Tort Liability

This is not to say that government employees, in their efforts to act

'8 Id. at 409-11 (internal citations and quotation marks omitted).

cient compensation to account for the risk of liability that employees

cerned about the risks of uncompensated loss could simply purchase

9 U.S. CONST. amend. V, cl.3.

II. THE POLITICS OF GOVERNMENTAL TORT LIABILITY

A. The Case Against Governmental Tort Liability

1. Theories of Tort Liability

law from mere loss-spreading through insurance by means of a con-

(1999); Heck v. Humphrey, 512 U.S. 477, 483 (1994).

16-19, 135-46 (1983).

VA. L. REV. 997, 1009-22 (1990).

2. The Inapplicability of the Conventional Theory to Government

are enjoyed by a majority of the population, compensation will never

of private corporations have the benefit of liability limited to their in-

political dividends in the public sector. Thus, after Levinson's attack,

,0 See supra text accompanying note 88.

ognized constitutional tort liability as we know it today in Monroe v.

Ramparts police corruption scandal. 37 What is more, only a tiny frac-

unless the government offered compensation amounting to effective

B. The Politics of GovernmentalDamages Liability

" See supra text accompanying note 91.

Consider the following account of political behavior as it relates to

.. See, e.g., POSNER, supra note 94, at § 19.3.

and spending policy is one all elected officials share, regardless of

2. Assessing the Theory

Many political controversies fall within the allocational arena. Loca-

available to elected officials available for allocational politics. For

'58 Harold J. Krent, PreservingDiscretion Without SacrificingDeterrence: Federal Government Liabil-